TWI269549B - Method, base station and terminal in a multiple-access multiple-input multiple-output communication system - Google Patents

Abstract

Techniques to achieve better utilization of the available resources and robust performance for the downlink and uplink in a multiple-access MIMO system. Techniques are provided to adaptively process data prior to transmission, based on channel state information, to more closely match the data transmission to the capacity of the channel. Various receiver processing techniques are provided to process a data transmission received via multiple antennas at a receiver unit. Adaptive reuse schemes and power back-off are also provided to operate the cells in the system in a manner to further increase the spectral efficiency of the system (e.g., reduce interference, improve coverage, and attain high throughput). Techniques are provided to efficiently schedule data transmission on the downlink and uplink. The scheduling schemes may be designed to optimize transmissions (e.g., maximize throughput) for single or multiple terminals in a manner to meet various constraints and requirements.

Description

1269549 (1) Kan, invention description (invention description should (4): the technical field, prior art, content, and method of the invention belong to the invention) BACKGROUND OF THE INVENTION

The present invention is generally directed to data communication, more specifically to a multi-directional proximity multiple input multiple output communication system. π·invention background

The communication system is widely developed to provide various communication types, such as female voices, materials, and so on. These systems can be multi-directional proximity systems that support communication (sequential or simultaneous) with multiple users by enjoying available system resources such as bandwidth and transmission power. The system can be architected in code division multi-directional handover (CDMA), time division multi-directional proximity (TDMA), crossover multi-directional proximity or other multi-directional proximity technology. In a wireless communication system (such as a cellular system, a broadcast system, a multi-channel multi-point distribution system (MMDS), and others), the RF modulated signal from a transmitter unit can reach a receiver unit via a plurality of propagation paths. . The characteristics of the propagation path usually change over time due to many factors, such as fading and multipath. Multiple transmit and receive antennas can be used to provide diversity to account for adverse path effects and improve performance. If the propagation path between the transmitting and receiving antennas is linearly independent (meaning that the transmission over one path does not form a linear combination of transmissions on other paths), at least in part to the extent that it is true, then correct The possibility of receiving a data transmission will increase as the number of antennas increases. As the number of transmit and receive antennas increases, diversity increases and performance improvements are generally achieved. 1269549

A multiple input multiple output (ΜΙΜΟ) communication system uses multiple (Ντ) transmit antennas and multiple (Nr) receive antennas for data communication. One of the multiple input multiple output channels formed using Ντ transmission and nr receive antennas can be decomposed into Nc independent channels, where S min {NT, NR}. Each Nc independent channel represents a spatial subchannel of the channel and corresponds to a range. If additional ranges are created by using multiple transmit and receive antennas, the system will provide improved performance (such as increased transmission capacity). The resources of a known communication system are typically limited by various control bundles and requirements' and other practical considerations. However, the system may need to support many terminals, provide various services, achieve certain performance goals, and more. Therefore, there is a need for a multi-directional proximity system that can flexibly operate and provide improved system performance. SUMMARY OF THE INVENTION A feature of the present invention provides techniques for enabling better utilization of available resources (e.g., transmit power and bandwidth) and robust performance for uplink and downlink within a wireless communication system. These techniques facilitate the use of a communication system (such as a multi-directional proximity system, using QFDMi) in a system, a proximity system (such as a CDMA, TDMA or FDMA system), an OFDM system, or any combination of the above. MIM〇 system, etc.). One feature provides techniques to process the bedding material based on channel state information prior to transmission to more closely match data transmission and channel capacity. With the adaptation of the transmission process, the coding and modulation scheme for data transmission can be selected according to the characteristics of the communication channel (CSI can be quantized by channel state information (CSI). The CSI can be at a receiver unit (such as a terminal) The decision is sent to a transmitter (3) 1269549 (such as a base station). The transmitter unit can then adjust the coding and modulation of the data transmission according to the reported c SI.

In another feature, techniques are provided to process a data transmission received via a multiple antenna at a receiver unit. Various receiver processing techniques are described herein as including a channel correlation matrix inversion (CCMI) technique, minimum mean square error (MMSE) technique, MMSE linear equalizer (MMSE-LE) technique, and a decision feedback equalizer. (DFE) technology and a continuous elimination receiver processing technique. These receiver processing techniques can be advantageously used in conjunction with adaptive transmission processing to achieve south performance. In yet another feature, the technology is provided to cells within the operating system to further increase the spatial efficiency of the system. The power transmitted over the downlink and/or uplink via adaptive re-use schemes and power snagging can be limited to an architectural approach to reduce interference, improve coverage, and maintain high throughput. In yet another feature, techniques are provided to efficiently schedule data transfers on the downlink and uplink. These scheduling schemes are designed to optimize single or multiple terminal transmissions (eg, maximize throughput) to meet various constraints and requirements (eg request requirements, load, fairness criteria, data transfer rate capabilities) , channel conditions, etc.). Some of the properties of the system (such as multi-user changes, receiver processing techniques, etc.) are also available to provide improved performance. The above and other features, specific embodiments and features of the present invention are described in further detail below. The present invention further provides a method, a transmitter unit, and a transceiver unit that can implement various features, embodiments, and features of the present invention. 1269549 _ (Don't talk about the hard page base station, terminal, system, device, program product, etc. The features, characteristics, and advantages of the present invention will become more apparent from the detailed description of the appended claims. 1 is a multi-directional proximity communication system diagram that can implement various features and embodiments of the present invention; FIGS. 2A and 2B are a base station and a second terminal for downlink and uplink data transmission, respectively. Figure 3A is a block diagram of a transmitter unit having a specific embodiment according to a portion of the available CSI adjustments; Figure 3 is a specific embodiment of the method for adjusting the processing according to the selected channel rotation. Block diagram of a transmitter unit; FIG. 3C is a block diagram of a transmitter unit having a specific embodiment in accordance with an overall CSI adjustment; 3D is a block diagram of a transmitter unit having a specific embodiment for independently encoding and modulating each group of transmission channels; FIG. 3 is a transmitter having a specific embodiment for independently processing data for each frequency subchannel of OFDM. Figure 4A is a block diagram of a specific embodiment of an RX ΜΙΜΟ/data processor in a receiver unit; Figure 4 Β, 4C, 4D and 4Ε are respectively capable of implementing CCMI technology, MMSE technology, DF Ε technology And the four implementations of the continuous elimination of the receiver processing technology 1269549 瞵 _ _ Continuation (5) Example of the space-time processor block diagram. Figure 4F is a channel ΜΙΜΟ / data processor specific in a receiver unit Figure 4 is a block diagram of a specific embodiment of an interference canceller; Figure 5 is a flow chart showing the technique of continuously erasing the receiver processing; Figure 6 is a diagram showing the system in accordance with a number of reuse modes. An example of a cumulative distribution function (CDF) of a signal to noise ratio achieved by a terminal;

6B shows an example of a CDF that achieves a signal to noise ratio (SNR) in a cell within a cell in a single cell reuse mode; FIG. 6C shows a graph of a specific embodiment of resource partitioning and arrangement of a three cell reuse mode, 7 is a flow chart of a specific embodiment of a program for adapting to reuse planning; FIG. 8 is a flow chart of a specific embodiment of a program for data transmission according to a rights scheduling terminal; FIG. 8B is a channel for specifying a terminal according to rights. Flow chart of a specific embodiment of the program;

8 is a flow chart of a specific embodiment of a procedure for upgrading a terminal to a preferred channel according to the authority; FIG. 9 is a program for performing downlink and uplink data transmission time intervals respectively for the A and 10 A scheduling terminals. FIG. 9B is a flow chart of a specific embodiment for specifying a transmitting antenna to a terminal using a maximum-maximum standard for downlink data transmission; and FIGS. 9C and 10B are scheduling a set of Ντ highest authority. A specific implementation of the procedures for the downlink and uplink data transmission time intervals of the terminal respectively - 1269549 · 丑 页 (6) Example flow chart; Figure 1 1 A shows each terminal in many operating modes The average downlink flux of the system between the four transmit antennas and the four receive antennas; Figure 1 1 shows the average uplink flux for four receive antennas with various numbers of single antenna terminals; and Figure 11C shows the simulation with 1, 2 and 4 transmit the cell flux of the cell network simultaneously transmitted by the antenna terminal.

DETAILED DESCRIPTION OF THE INVENTION I. Overall System Figure 1 shows a diagram of a wireless communication system 100 that supports a number of users and is capable of implementing embodiments of the present invention. System 100 provides communications for a number of coverage ranges 102a through 102g, each of which is serviced by a corresponding base station 104 (which may represent a proximity, Node B, or some terminology). The definition of the coverage of each base station can be, for example, the extent to which the terminal can reach a specific level of service (Go S). Base stations and their coverage are often referred to collectively as "cells."

As shown in Figure 1, the various terminals 106 are diversityd in the system, and each terminal can be a fixed (i.e., a fixed point) or an active end. Each terminal can communicate with one or more base stations on the downlink and/or uplink at any time, depending on whether it is powered on, whether it is "soft handover" or not. The downlink (forward link) represents the transmission from the base station to the terminal, and the uplink (reverse link) represents the transmission from the terminal to the base station. In Fig. 1, base station 104a communicates with terminal unit 106a, base station 104b communicates with terminals 106b, 106c and 106d, base station 104c and terminal sets 106e, 106f -11 - 1269549

(7) Communicate with 106g, etc. System 100 can also be designed to implement any number of standards and is designed for use in CDMA, TDMA, FDMA, and other multi-directional proximity systems. The cDMa standard includes IS_95, Cdma2000, IS-856, W-CDMA and TS_CDMA standards, while the TDMA standard includes the Global System for Mobile Communications (GSM) standard. These standards are known techniques and are incorporated herein by reference to the standards. The system 100 can be a multiple input multiple output (MIMO) system that uses multiple (Ντ) transmit antennas and multiple (Nr) receive antennas for data transmission. The multiple input multiple output channels formed by the % transmit and nr receive antennas can be decomposed into Nc independent channels, where nc S min {NT, Nr}. Each Nc independent channel also represents a spatial subchannel of the ΜΙΜΟ channel. The ΜΙΜΟ system provides improved performance (such as increased transmission capacity) if space sub-channels generated by multiple transmit and receive antennas are available. System 100 can selectively or additionally use orthogonal frequency division multiplexer (OFDM), which effectively divides the operating bandwidth into a majority (Nf) frequency subchannel (i.e., frequency bin). At each time slot (which is a specific time interval depending on the bandwidth of the frequency sub-channel), a modulation symbol can be transmitted on the NF frequency sub-channel. System 100 can be operated to transfer data via a number of "transport" channels. For systems that do not use 〇FDM, they usually have only one frequency subchannel and each spatial subchannel can be used as a transit channel. For the sub-channels of each frequency sub-channel using the OFDM system, the two sub-channels can be regarded as transmission channels. And for an OFDM system that does not use ΜΙΜΟ, there is only one spatial subchannel and each frequency sub-pass

1269549 _ (8) I send a note that the renewal can be used as a transmission channel. The following channels and subchannels can be supported by the system: • Channel-one transmission unit, which can be a time slot in a TDMA system, a frequency subchannel in an FDMA or OFDM system, or a coding channel in a CDMA system; Channel - a radio frequency propagation channel between the transmitting and receiving antennas; • a transmission channel - a spatial subchannel, a frequency subchannel or a spatial subchannel spatial subchannel in which separate data streams can be transmitted; • spatial subchannels - An independent channel formed by the spatial extent of the communication channel between the transmit and receive antennas; and • a frequency subchannel - a frequency bin within the OFDM system. Multiple antenna systems are used at both the transmitter unit and the receiver unit (i.e., NR X Ντ ΜΙΜΟ) - an efficient technique for enhancing the capacity of a multi-directional proximity system (i.e., cellular, PCS, LAN, etc.). Using ΜΙΜΟ, a transmitter unit can transmit multiple independent streams to a single or multiple receiver unit over the same communication channel by the spatial extent of the communication channel that couples the transmit and receive antennas. System 100 can be designed to support most modes of operation. In the system, each base station can be equipped with multiple transmit and receive antennas for data transmission and connection.

A is received, and each terminal can be equipped with a single transmit/receive antenna or multiple transmit/receive antennas for data transmission and reception. The number of antennas used for each type of terminal depends on various factors, such as services that are supported by the terminal (such as voice, data, or both), cost limits, management restrictions, and security issues - 13 - 1269549 _ (9) Send _ nickel _ continuation page and so on. Table 1 summarizes the operational mode matrix of the supportable system 100. Table 1 Transmitting Antenna Receiving Antenna 1 Nr 1 SISO SIMO Ν τ MISO ΜΙΜΟ The following briefly describes the operating modes of Table 1:

• SISO (Single Input / Single Output) - This RF link features a single transmit antenna and a single receive antenna. • SIMO (Single Input / Multiple Output) - This RF link features a single transmit antenna and multiple receive antennas. This mode of operation is available for diversity reception. • MISO (Multiple Input / Single Output) - This RF link features multiple transmit antennas and a single receive antenna. This mode of operation can be used for diversity transfers.

• ΜΙΜΟ (Multiple Input/Multiple Output) – This RF link is characterized by multiple transmit antennas and multiple receive antennas. When using ΜΙΜΟ, System 100 can be further designed to support the following modes of operation: • Diversity only - Use multiple transmit and receive antennas (i.e., transmit and receive diversity) to achieve a highly reliable transmission of a single data stream. • Spatial multiplex processing, single user (single user mode) - use multiple transmit and receive antennas to make a single terminal up to -1269549 by utilizing multiple parallel transmission channels generated by the spatial extent of the communication channel (丨〇) To high data transfer rate. Lu processing, most users (mostly use the MTMO mode to use multiple transmit and receive antennas to provide communication with most terminals currently on the same channel. • • Use multiple transmit and receive antennas to provide the same channel as currently SIMO communicates with most terminals in combination with ΜΙΜΟ. The above mode of operation can be considered as a subtype of ΜΙΜ〇 mode.

The specific mode of operation supported by each base station and each terminal depends in part on the number of transmit and receive antennas available at the base station and the terminal. A base station equipped with an antenna transmission antenna and multiple receiving antennas can support the operation modes listed above. A terminal can be designed to have any number of transmit and receive antennas. On the downlink, a terminal with a single receive antenna (if designed for voice services) can support SIS and MIS modes. Terminals with multiple receive antennas can support SIMO and ΜΙΜΟ mode. Some types of transmit diversity (ie, MIS 0) can be utilized for partial transmission of a single receive antenna terminal. In the uplink, a single transmit antenna terminal can support SISO and SIMO modes, while a multiple transmit antenna terminal can support MISO and ΜΙΜΟ mode. Spatial multiplex processing in a multi-directional proximity network combined with the space multiplex mode provides considerable system resiliency and further supports mixed-mode terminals. System settings for the downlink and uplink may vary by various factors, such as different service requirements, cost constraints, and the capabilities of different types of terminals. -15- 1269549

(ii) Multiple parallel channels are supported in a multi-user mode, where each such channel can be used as SIMO, ΜΙΜΟ or some combination. In the downlink, multiple transmit antennas at the base station can use parallel transmission channels to send data to different terminals. In this case, each terminal can utilize the multiple receive antennas to coordinate spatial processing to exclude signals from other terminals and demodulate its own signals. In the uplink, the base station's receiver unit uses multiple receive antennas to coordinate spatial processing to separately demodulate the transmissions from their respective terminals.

lose.

The multi-user mode is similar to spatially differentiated multi-directional proximity (SDMA). According to SDMA, "space signing" with different terminal authorities can be utilized to allow multiple terminals to operate simultaneously on the same channel. A space signature constitutes the complete radio frequency characteristic of the propagation path between the transmitting antenna and the receiving antenna. On the downlink, space signing can be derived at the terminal and then reported to the base station. The base station can then process these space signatures to select the terminal for data transmission on the same channel, and derive a common "orthogonal" operational vertical amount for each individual data stream for transmission to each selected terminal. In the uplink, the base station can derive space signing for different terminals. The base station can then process the space signing procedures for scheduling terminal data transfers and further process the transmissions by the selected terminals to demodulate the respective transmissions. If the terminal is equipped with multiple receiving antennas, the base station does not need to sign the space of the terminal to obtain the benefit of SDMA. The base station needs only after the terminal demodulation, a small amount of information from the terminal indicating the "post-processing" SNR and the connection from the base station transmit antenna signal. The SNR estimation program can be implemented by -16- (12) 1269549

The relay transmissions from the base stations are carried out periodically - the pilot, as explained below. In the case of the second uplink, the base station can be allocated by subscribing resources and users (for example, on a request basis). When the resource allocation to #^ τ is provided to the user via the control channel to indicate the use of the <special operation mode. ^ Sex production, the material 'system can be based on system load and / or some, and a variety of operational parameters (such as operating mode, pass:, :: transmission: rate, transmit antenna, transmit power, etc.)... It: Resentment, as explained below. Soil 2. Block Diagram Figure 2A is a block diagram of the station (10) and the two-terminal machine 106 in the system for the downlink data transmission. At the base station 1〇4, the data source 2〇8 provides the bedding material (ie, the information bit) to a transmission data processor 21〇. For each transmit antenna, the τχ data is stored in the ii 2iq(1) according to the special code: the code is decoded, and (7) the interleaved bits are interleaved (ie, recombined) according to a specific interleaving scheme and (3) are mapped. The meta-modulation symbol is used by _ or multiple transmission channels selected for data transmission. This code can increase the reliability of the data transmission. The interleaving provides time diversity for the coding bits to use 'allowing the data to be transmitted according to the average transmission of the transmission channel, against the case of strong and weak, removing the correlation between the coding bits used to form the modulation symbols, and if encoded The frequency diversity can be further provided when the bit is transmitted through the multiple frequency subchannel. In one feature, encoding and modulation (i.e., symbol mapping) can be performed in accordance with control signals provided by controller 230. The ΜΙΜΟ processor 220 receives and de-multiplexes the modulating symbols from the τχ data processor 2 10 and provides them for each transmission channel (eg, -17-1269549 (13). A modulation symbol stream of the antenna), each time slot is a modulation symbol. The processor 220 can further pre-adjust the modulation symbols for each of the transmission channels if a complete C S I (e.g., a channel response matrix) is available. The complete and complete CSI processing will be described in further detail below. If OFDM is not used, the processor 220 will provide a modulation symbol and stream for use by each transmit antenna for data transmission. And if OFDM is employed, the processor 220 provides a modulated symbol vector stream for use by each of the transmit antennas used for data transmission. And if a full CSI process is performed (described below), the processor 220 provides a pre-adjusted modulated symbol stream or an adjusted modulated symbol vector for each antenna used for data transmission. Each data stream is received and modulated by a respective modulator (M〇D) 222 and transmitted via the associated antenna 2 24 . At each of the terminals 1 to 6 where the data transmission arrives, one or more antennas 25 2 receive the signals transmitted thereto, and each of the receiving antennas provides the received signals to respective demodulator (DEMOD) 254. Each demodulator (or front end unit) 254 performs a process complementary to modulator 222. The modulated symbols from each of the demodulators 254 are then provided to a receive (RX) MIM/data processor 26 and then processed to _back k for transmission to one or more data streams of the remote terminal. The RX MIM0/data processing state 260 implements processing complementary to the data processor 21〇 and the τχ MIM processor 22, and provides decoded data to a data slot 262. The processing via terminal 106 will be described in further detail below. _ At the active terminal 106, the RXMIM/data processor 26 further estimates the downlink and provides channel status information (csi) (eg, post-processing SNR or channel gain estimate) indicating the predictive chain. Road situation. Controller 27〇 -18- 1269549

(14) Receive and further convert the downlink CSI (DL CST) into some other type (eg data rate, coding/modulation scheme, etc.). A TX data processor 280 then receives and processes the downlink CSI and provides downlink CSI (either directly or via a τχμίο processor 282) to one or more modulators 2 5 4 ° mutator(s) 254 further adjusts the processed data and transmits the downlink CSI back to the base station 1〇4 via an opposite channel. The downlink CSI can be reported by the terminal using various signaling technologies as explained below. At base station 104, the transmitted feedback signal is received by antenna 224, demodulated by demodulator 22, and then provided to an rx/data processor 24 0 RX/data processor 240. The data processor 280 is processed in a complementary manner to the processor 2 8 2 (if used), and then the delivered CSI is returned to the controller 230 and the scheduler 234. The scheduler 2 3 4 uses the downlink C S I to perform many functions, such as (1) selecting the best terminal unit for data transmission and (2) designating the available transmission antenna to the selected terminal. Scheduler 2 3 4 or controller 203 further uses the reported downlink CSI to determine the coding and modulation scheme to be used by each transmit antenna. Scheduler 2 3 4 can be programmed as described below to achieve high throughput and/or according to other performance criteria or metrics. Figure 2 is a block diagram of a base station 1〇4 and a second terminal 1 0 6 for uplink data transmission. At the terminal 106 scheduled for uplink data transmission, a data source 278 provides data to the data processor 280, which encodes, interleaves, and maps the data into modulation symbols. If multiple transmission antennas are used for data transmission, the processor 282 receives and further processes the modulation symbols to provide a modulated symbol stream 'pre-adjusted modulation symbols, -19-1269549

(15) Adjust the pay vector or pre-adjusted modulation symbol vector — data transmission. The data streams are then received and modulated by respective modulators 254 and transmitted via an associated antenna 252. At base station 104, a plurality of antennas 224 receive the transmitted signals and each antenna provides the received signals to respective demodulators 2 2 2 . Each demodulator 2 2 2 performs a process complementary to the modulator 254. The modulation symbols from all of the demodulator 222 are then provided to the RX ΜΙΜΟ/data processor 240 and then processed to reply to the data stream transmitted by the terminal device being queued. The rx μιμ / data processor implements complementary processing with the data processor 28 and the processor 2 82 (if used), and then provides decoded data to a data slot 242. For each terminal 106 that needs to transmit at an upstream transmission interval, the RX/ΜΙΜΟ processor 24 further estimates the channel conditions for the uplink and derives the uplink CSI (UL CSI), which will be provided to the control. Device. The scheduler 234 can also receive and use the uplink csi to reduce (7) the specific processing of the signal from the selected terminal to be used by each of the transmitted antennas of each of the scheduled terminals. For each transmission interval, the scheduler 234 provides a terminal, which is selected for the data transmission, and a designated two-way: the transmission parameters of the field machine. Each of the scheduled transmissions, terminals, and parameters of the terminal may include the data transmission rate to be used and the coded iron, 'spring' transmission TX data processor 210 receiving and processing the uplink T million case . The display is routed to one or more of the modulators 222 schedules: and the indications are further in effect. The modulator 222 adjusts the processed data and transmits the uplink via the wireless port -20- 1269549 (16)

Schedule to the terminal. The uplink schedule can be sent to the terminal using various signaling and messaging techniques. At each active terminal 106, the transmitted signal is received by antenna 252, demodulated by demodulator 254, and then supplied to RX ΜΙΜΟ / data processor 260. The processor 26 is implemented and processed by the processor 220 and the data processing

The processor 210 is complementary to the process and then replies to the uplink schedule for the terminal (if needed) and then to the controller 27 to control the uplink transmission via the terminal. The scheduler 234 shown in Figures 2A and 2B is located at the base station 1〇4. In other implementations, scheduler 234 may be located within some other component of system 1 (e.g., a base station controller coupled to and interfaced with certain base stations). II· Transmitter unit

A helium system that would provide improved performance if an increased range produced by multiple transmit and receive antennas is used. Increasing system efficiency and performance is possible if the transmitter unit has a C S I that describes the transmission characteristics from the transmit antenna to the receive antenna (although this is not absolutely necessary). C S I can be classified as "complete CSI" or "partial CSI". The complete C SI includes sufficient features (e.g., amplitude and phase) across the entire system bandwidth (i.e., frequency subchannels) for the propagation path between the transmit-receive antenna pairs within the (Ντ X NR) ΜΙΜΟ matrix. The complete C SI processing means that (1) the channel characteristics can be used for both the transmitter unit and the receiver unit, and (2) the transmitter unit derives the natural mode of the channel (described later) and decides that it will be transmitted in the eigen mode. Modulation symbol, linear pre-modulation (filtering) the modulation symbol and transmitting the pre-adjustment modulation symbol, and (3) receiver unit according to channel special - 21 - (17) 1269549 micro-implementation processing complementary to linear transmission processing (If the space is matched too much), it is used to guide each transmission channel (that is, each eigenmode) (space matching factor. The complete CSI processing is based on the eigenvalue of the channel (described later). Each of the transmission channels uses a suitable encoding and modulation scheme to derive the modulation symbols. Part (3) may include, for example, a signal-to-noise plus interference ratio (SMR) of the transmission channel. The SNR of the specific transmission channel may be Detecting - the stream of funds or the directors transmitted on the trajectory is derived. Partial (3) burying means that it is processed according to the appropriate coding and variants for each transmission channel selected according to the 81^ han value of the channel. In On the downlink and uplink, either complete or partial (3) can be used to adjust the operating parameters of various systems. On the downlink, the terminal can derive the SNR for each transmission channel and report via the opposite channel. Downlink CSI to the base station. The base station will then use this information to schedule the downlink transmission to the 2 ray, and determine the channel and antenna designation, mode of operation, data rate and transmit power to be used. On the link, the base station can derive the SNR about the respective terminal and then use this information to schedule the uplink transmission. Relevant information (such as scheduling, ZECO transmission rate, coding and modulation scheme, transmission power, etc.) can be It is transmitted to the affected terminal through the control channel on the downlink. L [Partial-CS丄Processing Transmitter Single Figure 3 Block diagram of the eight-plane 〇 transmitter unit 纟 实施 , , A specific embodiment of the transmitter portion of the base station 1〇4 or the terminal unit 1〇6 in Figures 2A and 2B. The transmitter unit 300a can be based on the available portion of csi (Example -22- 1269549)

(18) If it is reported by the receiver unit, adjust its processing. The transmitter unit 3A includes (1) a TX data processor 2 1 Oa, which can receive and process information bits to provide modulation symbols, and (2) a processor 220a that can demodulate the variable symbols. The multiplex processing is used for the Ντ transmission antenna.

TX data processor 210a is a specific embodiment of TX data processors 210 and 208 in Figures 2A and 2B. In the particular embodiment illustrated in FIG. 3A, TX data processor 210a includes an encoder 3 12, a channel interleaver 3 14 and a symbol mapping component 3 16 . Encoder 3 1 2 receives and encodes information bits in accordance with a particular coding scheme to provide coded bits. The coding scheme may comprise a convolutional code, an accelerating code, a block code, a cyclic redundancy code (CRC), a co-colcode or other code or a combination of the above. The channel interleaver 3 1 4 interleaves the encoded bits in accordance with a particular interleaving scheme to provide diversity. The symbol mapping component 3 16 maps the encoded bit into a modulation symbol for one or more transmission channels for transmitting data.

Although not shown in Figure 3A for simplicity, a preamble (e.g., data of a known pattern) can also be encoded and processed with multiplexed information bits. The preamble can be transmitted in all transmission channels or subsets thereof for transmitting information bits (e.g., in one-time multiplex processing (TDM) or one-code multiplex processing (CDM)). The preamble can be used at the receiver to implement channel estimation, frequency and timing estimation, correlation data demodulation, and more. As shown in Figure 3A, the encoding and modulation can be adjusted based on the portion of the CSI available, as reflected in the encoding and modulation control. In a specific embodiment, the adaptive coding is achieved by using a fixed base code (eg, a ratio of 1 / 3 acceleration), and when the SNR is supported by the transmission channel for transmitting data -23 - 1269549 _ Continued on page (19), adjust the penetration to achieve the required coding rate. For this coding scheme, penetration can be implemented after channel interleaving. In another embodiment, different coding schemes may be used depending on the portion of the C S I that is available (e.g., each data stream may be encoded according to a separate code).

For each transmission channel, symbol mapping component 3 16 can be designed to cluster interleaved bytes to form non-binary symbols, while mapping each non-binary symbol to a particular modulation scheme (eg, QPSK) that should be selected for the transmission channel. A point in the signal group of M-PSK, M-QAM or some other scheme. Each mapped signal point corresponds to a modulation symbol. The number of information bits that can be transmitted for each particular level of performance (e.g., one percent packet error rate (PER)) depends on the SNR of the transmission channel. Therefore, the coding scheme and modulation scheme for each transmission channel can be selected based on the available partial CSI. The channel interleaving can also be adjusted according to the available portion C S I as shown by the dashed line entering the encoding control of block 314.

Table 2 lists the various combinations of coding rates and modulation schemes that can be used for many SNR ranges. The bit rate supporting each transmission channel can be achieved by any of the possible coding rates and modulation schemes. For example, each modulation symbol-information bit can be achieved using (1) 1/2 and QPSK modulation coding rate, (2) 1/3 and 8-PSK modulation coding rate, (3) 1/ 4 and 16-QAM coding rate, or some combination of other coding rate modulation schemes. In Table 2, QPSK, 16-QAM and 64-QAM are used to list the SNR range. Other modulation schemes such as 8-PSK, 32-QAM, 128-QAM, etc. may also be included and fall within the scope of the invention. -24- 1269549 _ (20) I send instructions to continue: page 2 SNR information bits / modulation coded bits / code rate range symbol number symbol symbol number 1.5 - 4.4 1 QPSK 2 1/2 4.4 - 6.4 1.5 QPSK 2 3/4 6.4-8.35 2 16-QAM 4 1/2 8.35-10.4 2.5 16-QAM 4 5/8 10.4-12.3 3 16-QAM 4 3/4 12.3 - 14.15 3.5 64-QAM 6 7/ 12 14.15 -15.55 4 64-QAM 6 2/3 15.55 -17.35 4.5 64-QAM 6 3/4 >17.35 5 64-QAM 6 5/6

The modulation symbols from TX data processor 210a will be provided to a TX processor 220a, which is a specific embodiment of processors 220 and 282 in Figures 2A and 2B. In a processor 220a, a demultiplexer 3 24 processes the received modulated symbol demultiplexing into a plurality of (Ντ) modulated symbol streams for each antenna to transmit the modulated symbols. Each modulated symbol stream will be provided to a respective modulator 2 2 2 . Each modulator 2 2 2 converts the modulated symbol into one or more analog signals, and further amplifies, filters, quadrature modulates, and upconverts the signal to generate a wireless bond suitable for transmission via the associated antenna 2 24 A modulated signal transmitted. If the number of spatial subchannels is less than the available transmit antenna (ie, Nc < Ντ), then various schemes can be used for one data transmission. In one arrangement, the Nc modulation symbol stream is generated and transmitted with a subset of available transmit antennas (i.e., Nc). The remaining (NT - Nc) transmit antenna will not be used for data transmission. In another approach, the additional freedom provided by the additional transmit antenna (NT - Nc) will be used to improve the reliability of data transmission. For this solution, each one or more -25 - 1269549 is said to be purchased (21) streams can be encoded, possibly interleaved, and then transmitted through multiple transmit antennas. The use of multiple transmit antennas for a given data stream can increase diversity and improve reliability for unfavorable path effects. 2. ΜΙΜΟ transmitter unit with selective channel rotation

Figure 3 is a block diagram of a specific embodiment of a tethered transmitter unit 300b that can process data in response to selective channel rotation. To simplify data processing at both the transmitter and receiver units, a common coding and modulation scheme can be used for all transmission channels selected for data transmission. In this case, the transmitter unit will encode the data using a single (e.g., convolution or acceleration) code and decoding rate, and then map the resulting coding bits to the modulation symbols using a single (e.g., PSK or QAM) modulation scheme. To support this single encoding and modulation scheme, the transmit power levels for each selected transmission channel can be set or adjusted to achieve a particular SNR at the receiver unit. The power control can be achieved by "swinging" the selected transmission channel and appropriately distributing the available transmit power to all selected channels.

If the transmit power for all available transmission channels is equal and the noise variation σ2 of all channels is equal, the SNR of the relevant transmission channel (j, k) is received, Y(j, k) can be expressed as: τϋΛ)

σ2ΝτΝΡ Equation (1) where Prx(j,k) is used for the received power of the transmission channel (j, k) (ie the jth space subchannel of the kth frequency subchannel), and the overall transmission available at the Ptx transmitter unit Power, and H(j,k) is the composite channel gain (if ΜΙΜΟ is not used, j=l, and if OFDM is not used, k=l). -26- 1269549

(22) The normalization factor β used to distribute the overall transmit power in the selected transmission wanted can be expressed as: β=~Στΰ^ Equation (2) Wide (/, too » decisions where hh is used to select transmissions An SNR threshold. As shown in equation (2), the normalization factor β is calculated from the SNR of all selected transmission channels (and the reciprocal of the sum). A similar received SNR is achieved for all selected transmissions. For the wheel channel, the modulation symbol of each selected transmission channel (J, k) can be weighted by the weighting amount W(j, k) of the channel snr, which can be expressed as: W{jfk) Equation (3) Each transmission channel The weighted transmission power can then be expressed as βΡα< Λ8 J,k), where /C/,4), other equations (4)

As shown in the equation (4), only the transmission channel of the received s NR big boundary value (ie Y(j, k) 2 Yth) will be selected. The selective channel rotation system is detailed in In May 2001, the application number is 〇9/860,274, and on June 14, 2001, the application serial number 09/881,6 10 and June 26, 2001, application serial number 09/892,379, all three titles are r The use of selective channel reversal for processing in the system is assigned to the assignee of the present application and is used by the reference party # < equal to SNR ^.曰中的+请的美国 $ <US Patent Patent Application is incorporated in the multi-channel system. -27- 1269549

(23) As shown in Fig. 3B, the transmitter unit 300b includes a τ χ data processor 2 coupled to a τ χ processor 220b. The data processor 21A includes an encoder 312, a channel interleaver 314 and a symbol mapping element 316, which operate as described above. TX data processor 210b further includes a symbol weighting component 318 that can be weighted by a respective weighting value to provide a weighted modulation symbol. The weighting value of each selected transmission channel can be determined as described above based on the S N R of the arrival of the channel and the SNR of the other selected transmission channels. The SNR threshold Yth can be determined as described in the aforementioned U.S. Patent Application Serial Nos. 09/860,274, 09/881, 61, and 9/892,379. 3. IMO Transmitter Army with Complete CS1 Processing Figure 3 is a block diagram of a specific embodiment of a C-Series transmitter unit 300c that can process data based on the complete CSI provided by the receiver unit. Transmitter unit 300c includes a TX data processor 210c coupled to a processor 220c. The data processor 21 Ob includes an encoder 312, a channel interleaver 314 and a symbol mapping element 3 16, which operate as described above. The processor 220c includes a channel buffer processor 322 and a demultiplexer 324. The channel ΜΙΜΟ processor 3 2 2 demultiplexes the received modulating symbol into a plurality of (Nc) modulated symbol streams, one of which is used for each spatial subchannel (ie, eigenmode) to transmit the modulating symbol . For the complete C S1 process, the channel ΜΙΜΟ processor 3 22 pre-adjusts the Nc modulation symbol in each time slot to generate the Ντ pre-adjusted modulation symbols as follows: , 丨, 12 12, ^1NC ^2 Μ = 彡21, e2Sc jVr ΝβΝτ2» eNTNc.

• b'· b2 M Equation (5) -28- 1269549

(24) where bi, b2, ..., and bNc are modulation symbols of spatial subchannels 1, 2, ..., Nc, respectively, wherein each Nc modulation symbol can use, for example, M-PSK, M-QAM, or some Other modulation schemes are generated; elements of a eigenvector matrix 1 relating to transmission characteristics from the transmit antenna to the receive antenna; and xi, x2, ... xNT are pre-conditioned modulation symbols.

The eigenvector matrix may be calculated by the transmitter unit or provided to the transmitter unit (e.g., by a receiver unit).

For complete CSI processing, each preconditioned modulation symbol (Xi) is used for a particular transmit antenna to represent a linear combination (weighted) modulation symbol for up to Nc spatial subchannels. The modulation scheme for each modulation symbol is selected based on the effective SNR of the natural mode and is proportional to the eigenvalue (described later). Each of the Nc modulation symbols used to generate each of the pre-conditioned modulation symbols can be associated to a different signal group. For each time slot, the Ντ pre-tuning modulation symbols generated by the channel ΜΙΜΟ processor 322 are demultiplexed and provided to the 调τ modulator 222 by a demultiplexer 3 24 . Complete CSI processing can be implemented depending on the available CSI and a subset of all or transmit antennas. The complete C S I process can also be used selectively and/or dynamically. For example, full C S I processing may be available for a particular data transmission system and not for some other data transmission system. Complete CSI processing can also be used under certain conditions, such as when the communication link has the appropriate SNR. 4. A transmitter unit with independent processing. Figure 3 is a block diagram of a specific embodiment of a D-system transmitter unit 300d, which can be -29 - 1269549

(25) Independently encode and modulate the data for each group of transmission channels based on the specific coding and modulation scheme selected for the group. In a particular embodiment, each group corresponds to a transmit antenna, and the transmit channels in each group may correspond to the transmit antenna < frequency sub-channel. In another embodiment, each group corresponds to a respective receiver unit to which a data transfer wheel 1 is directed. Typically, each group can include a number of transmission channels, with the data being encoded and modulated by a common coding and modulation scheme. Transmitter unit 300d includes a TX data processor 21〇d coupled to _τχ ΜΙΜΟ processor 220d. The data processor 21〇d includes a plurality of subchannel data processors 3 10a to 3 10t and a data processor 31 for each group of transmission channels for independent encoding and modulation. In the particular embodiment illustrated in Figure 3D, each data processor 310 includes an encoder 312, a channel interleaver 314, and a symbol mapping component 316 that operates as described above. In the particular embodiment illustrated in Figure 3D, the modulation symbols from each of the data processors 31 are provided to respective combiners 3 26 within the processor 220d. If the frequency sub-channels selected for a particular transmit antenna are included in each group, then the modulation symbols of the selected frequency sub-channels are combined to form a modulation symbol vector for each time slot, which is then provided to Respective modulators 222. The process of generating a modulated signal by each of the modulators 222 will be described later. In some other specific embodiments, the processor 220d can include a combiner and/or a demultiplexer to combine the modulation symbols and/or to demultiplex the modulation symbols to their appropriate modulators 222. 5. OFDM ΜΙΜΟ transmitter unit 图 Figure 3 Ε 发射 transmitter unit 3 0 0 e — block diagram of a specific embodiment, using -30 - 1269549

(26)

OFDM and the ability to process data independently for each frequency subchannel. In a τ χ data processor 210e, the information bit stream for each frequency sub-channel for data transmission will be provided to a respective frequency sub-channel data processor 340. Each data processor 320 is a respective frequency subchannel processing material for the 0 FDM system and can be implemented similar to the TX data processor 21A, 210b or 210d, or some other design. In one embodiment, the data processor 309 multi-channel allocation frequency sub-channel data stream becomes a plurality of data tributaries, and a data tributary stream is used for each spatial sub-channel selected for the frequency sub-channel. Each data stream is then encoded, interleaved, and symbol mapped to produce a modulation symbol for the data stream. The coding and modulation of each frequency subchannel data stream or each data stream can be adjusted according to the coding and modulation control signals. Each data processor 310 provides a modulated symbol stream of up to Nc for the Nc spatial subchannel selected for that frequency subchannel.

For systems using OFDM, the modulation symbols can be transmitted from multiple transmit antennas and on multiple frequency subchannels. In a processor 220e, the modulated symbol streams from each of the data processors 330 to Nc are provided to a respective spatial processor 323, which is controlled according to the channel and/or available for CSI processing. The modulated symbol received. If the full CSI process is not implemented, each spatial processor 332 can simply implement a demultiplexer (as shown in Figure 3A), or can implement a demultiplexer after a processor (Figure 3). As shown in C), if the complete CSI processing is implemented. For a system using OFDM, full CSI processing (i.e., pre-conditioning) can be implemented for each frequency sub-channel. Each spatial processor 3 3 2 multiplexes each time slot up to N c modulation symbol into -31-

1269549 (27) Use up to Ντ modulation symbols for transmission antennas selected for this frequency subchannel. For each transmit antenna, a combiner 314 receives the modulation symbols of up to the NF frequency subchannels selected for the transmit antenna 'combined with the symbols for each time slot to become the modulated symbol vector V, and provides the modulated symbol The vectors are to respective modulators 222.

The processor 220e thus receives and processes the modulation symbols to provide modulation symbols (Vi to VNt) of up to Ντ, a modulation symbol vector for use by each of the transmit antennas selected for data transmission. Each of the modulated symbol vectors V encompasses a single time slot, and each component of the modulated symbol vector V is coupled to a particular frequency sub-pass with a unique carrier on which the modulated symbol can be transmitted. FIG. 3E also shows a specific embodiment of OFDM modulator 222. The modulation symbol vectors Vi to VNt from the TX channel processor 220e will be supplied to the modulators 222a through 222t, respectively. In the embodiment illustrated in FIG. 3E, each modulator 222 includes an inverse fast Fourier transformer (IFFT) 340, a cyclic prefix generator 342, and an upconverter 344. IFFT 340 uses IFFT to convert each received modulated symbol vector into its time domain indication (which is referred to as an OFDM symbol). The IFFT 340 can be designed to implement the IFFT on any number of frequency subchannels (e.g., 8, 16, 32, ..., NF). In one embodiment, since each modulated symbol vector is converted to an OFDM symbol, the cyclic prefix generator 342 repeats a portion of the time domain indication of the OFDM symbol to form a "transport symbol" for a particular transmit antenna. The cyclic prefix ensures that the transmitted symbol maintains its orthogonality under the presence of multiple path delay spreads, thereby improving the effectiveness of preventing adverse path effects. The implementation of IFFT 340 and cyclic prefix generator 342 is a previously known technique, thus -32 - 1269549

(28) Not detailed here. The time domain indication from the cyclic prefix generator 342 (i.e., the transmitted symbols of the antennas) is then processed by the upconverter 34 4 (e.g., converted to an analog signal, shouldered, amplified, and over) to produce a tone. The signal is changed, which is then transmitted by a separate antenna 2 2 4 . An example of a system using a 0 FDM is disclosed in U.S. Patent Application Serial No. 09/532,492, filed on March 30, 2003, entitled "Using Multi-Tag Modulation for High Efficiency, High Performance Communication The system is hereby incorporated by reference to the assignee of the present disclosure. 〇FDM modulation is also described in a document titled “Multiple Carrier Modulation for Data Transmission: Its Transcendence Has Come,” by John AC Bingham in the May 1990 issue of IEEE Communications Hybrid. The manner of reference is incorporated herein. Figures 3A through 3E show examples of certain coding and modulation schemes that may advantageously be combined with full or partial CsI to provide improved performance (e.g., higher throughput). Some of these encoding and modulation schemes are further described in U.S. Patent Application Serial Nos. 09/826,48 1 and 09/956,449, both of which are entitled "Using Channel Status Information in a Wireless Communication System" And device". Applications filed on March 23, 2001 and September 18, 2001, respectively, and U.S. Patent Application Serial No. 09/854,235, filed on May 11, 2001, entitled "A Multiple Input Multiple Output (MIM) 〇) Method and apparatus for using channel status information for data processing in a communication system. Some examples of other coding and modulation schemes are described in U.S. Patent Application Serial No. 09/776,075, filed on Feb. 1, 2011. The title is "Encoding Scheme for Wireless Communication System". These applications are hereby incorporated by reference in its entirety in its entirety by reference in its entirety in its entirety in its entirety in Still other coding and modulation schemes can be used and are within the scope of the invention. 6. Mode of Operation Various modes of operation that can be used to apply adaptive transmitter processing techniques in accordance with the available c S I for a single and/or one of the F FDM systems are detailed herein. Some of the advantages will be outlined below.

In one mode of operation, the encoding and modulation of each transmission channel is selected based on the transmission capability of the channel, as indicated by the available C SI (e.g., SNR) for that channel. This approach provides performance improvements, particularly when used in conjunction with the continuous erase receiver processing techniques described below. When there is a large difference between the worst-case and best-condition transmission channels, the code can be selected to introduce enough redundancy to allow the receiver unit to revert to the original data stream. For example, the worst transmission channel can be coupled to a poor SNR at the receiver output. A forward error correction (FEC) code can then be selected to effectively allow the symbol to be transmitted on the worst condition transmission channel and correctly received by the receiver unit. When the transmitter has an SNR on each of the recovered transmitted signals, different encoding and/or modulation methods can be used for each transmitted signal. For example, depending on its S N R, a particular coding and modulation scheme can be selected for each of the transmitted signals such that the error rates associated with the transmitted signals are nearly equal. In this manner, the flux of the transmitted signals is governed by their respective SNRs, rather than the SNR of the transmitted signal by the worst conditions. In another mode of operation, the transmitter does not have a CSI for each transmission channel, but may have a single value indicating the average characteristics of all its transmission channels - 34 - 1269549 (30) _ Wei Lai Continuation Page:: Yi: Face (for example Average SNR), or some information that may indicate which transmit antenna will be used for data transmission. In this manner, the transmitter can utilize the same coding and modulation scheme on all transmit antennas for data transmission, which can be a subset of the available transmit antennas.

If the coding and modulation scheme is used for all or most of the transmitted signals, the recovered transmitted signal with the worst SNR will have the highest decoding error rate. This greatly limits the performance of the system because the coding and modulation schemes are selected such that the error rate for the worst-case transmitted signals meets the overall error rate requirements. To improve efficiency, additional receive antennas can be used to provide an improvement in error rate performance over the transmitted transmitted signal. By using a receiving antenna having a larger number of transmitting antennas, the error rate of the first reply transmitted signal has a degree of variability (NR - NT + 1), and the reliability is increased.

In yet another mode of operation, the transport stream "circulates" across all available transmit antennas. This approach improves the SNR statistics for each of the recovered transmitted signals because the transmitted data is not about the worst conditional transmission channel but for all transmission channels. The decoder for a particular data stream is effectively represented by a "soft decision" which represents the average across all transmit-receive antenna pairs. This mode of operation is further described in European Patent Application Serial No. 99302692.1, entitled "Wireless Communication System with Space Time Architecture Using Multiple Component Antennas at Transmitter and Receiver 2" and is incorporated herein by reference. 7. Transmitting Antennas The transmitting antenna groups at the base station can be actually different "opening" groups, each of which can be used to directly transmit their respective data streams. Each opening may be -35- 1269549 (31) formed by a collection of one or more antenna elements distributed in two (e.g., actually in a single location or distributed in multiple locations). Alternatively, the antenna apertures may be preceded by one or more (fixed) beamforming matrices, wherein each matrix is used to synthesize an antenna beam of a different set than the set of apertures. In this case, the above description of the transmitting antenna can be equally applied to the switching antenna beam.

Most fixed beam forming matrices can be defined in advance, and the terminal can evaluate the post processing for each possible matrix (antenna beam group) and then send the S N R vector to the base station. Usually different types of converted antenna beams can achieve different performance (i.e., post-processing SNR), which can be reflected in the reported SNR vector. The base station can then schedule and antenna design (using the reported SNR vectors) for each possible beamforming matrix, and select a particular beamforming matrix and a set of terminals and their designated antennas to achieve the best available resources. use.

The use of a beamforming matrix provides additional flexibility in scheduling the terminal and can further provide improved performance. In the example, the following states are well suited for beamforming conversion: • The connectivity within the channel is so high that a small amount of data stream can achieve optimal performance. However, transmission only with a subset of the available transmit antennas (and only with its associated transmit amplifier) will result in a smaller transmit power. A conversion can be selected to use all of the transmit antennas (and their amplifiers) for transmitting the data stream. In this case, a higher transmission power can be achieved for transmitting the data stream. • Substantially diversity terminals can be separated to some extent by their location -36- 1269549 嚷 明 Μ Μ (32). In this case, the function of the terminal can convert the horizontally split aperture into a set of beams pointing to different azimuth angles by a standard F F Τ type conversion. III. Receiver unit

A feature of the present invention provides a technique for processing a received signal in a system to recover the transmitted data, and estimating the characteristics of the channel. The C S I indicating the channel characteristics is then reported back to the transmitter unit and used to adjust signal processing (e.g., encoding, modulation, etc.). In this way, high performance can be achieved depending on the determined channel conditions. If the number of receive antennas equals or exceeds the number of transmit antennas (i.e., Ντ of Nr), several receiver processing techniques can be used for single-user and multiple-user modes. These receiver processing techniques can be clustered into two main types: • Space and space time receiver processing techniques (also known as equalization techniques), and

• “Continuous Invalidation/Equivalent and Interference Cancellation” Receiver Processing Technology (or “Continuous Elimination” Receiver Processing Technology). In general, spatial and spatial time receiver processing techniques desirably separate transmission signals at the receiver unit, and each separated transmission signal can be further processed to recover data included in the signal. The continuous cancellation receiver processing technique expects (sequentially) to reply to the transmitted signal and eliminate the interference caused by each of the recovered signals, such that the later recovered secondary signal experiences less interference and a higher SNR. Continuous elimination of receiver processing techniques generally outperforms (ie, has greater throughput) spatial/spatial time receiver processing techniques. -37- 1269549 Hairpin: Jihe (33) The use of continuous elimination receiver processing technology can be limited to certain situations. In particular, interference cancellation is only effective if the interference generated by a reply signal can be accurately estimated, which would require zero error detection (i.e., demodulation and decoding) of the reply signal.

On the downlink, if a single user port mode is used and the terminal is equipped with multiple receive antennas, the performance cancellation receiver processing technique will be available. If multiple user mode is used, a capable terminal can use space/space time receiver processing (ie, no persistent cancellation). Therefore, the terminal capable of the capability may not be able to reply to the transmission signal scheduled for another terminal (because the encoding and modulation method selected for the transmitted signal may be based on the processing SNR of other terminals) Interference with this transmitted signal cannot be eliminated.

For the downlink, the simplest is to use space/space time receiver processing techniques for all terminals when using the multiple user mode. After deriving the respective transmitted signals at the terminal, the processed SNR can be reported to the base station, which can then use the information to optimally schedule the terminal for data transmission, specify the transmit antenna to the terminal, and properly encode and modulate data. On the uplink, a single receiver unit at the base station replies with signals transmitted from one or more terminals, and the continuous cancellation receiver processing technique is generally applicable to single-user and majority-user modes. . In the single-user mode, the base station receiver unit derives the transmitted signals and processes the SNR, which can then be used for scheduling and coding and modulation. In the multi-user mode, the base station receives -38- (34) _ 1269549 to derive the driver terminal (ie 垤SN^, and ** owes H field and seeks the Bec transmitter). It can be used to select the rocker, the input, the 5, the good (., and the terminal for data transmission, and the coding and modulation method for each transmitting antenna. According to the characteristics of the 吣M0 channel, the π 彳 彳Receiver processing technology

Break can be divided into non-distributive or non-distributive 遒 遒 厣 厣 厣 MI MI MI MI MI , , , , , 千 千 千 千 千 千 千 千 千 千 千 千 千 千 千 千 千 千 千 千 千 千 千 千 千 千 千 千 千 千 千 千 千Sexual recession (such as the horizontal system has a different system bandwidth).

For the non-dispersive ΜΙΜ0 channel, linear spatial processing techniques (such as the channel association matrix inversion (CCM) technique, minimum mean square error (MMSE) technique and complete CSI technology, which are further elaborated below, can be used in demodulation = Processing the received signal before decoding. Other receiver processing techniques can also be used' and are within the scope of the invention. These spatial processing techniques can be utilized at the receiver unit to remove unwanted signals, or in the presence of other signals. The noise and interference maximize the reception of each constituent signal. The ability to effectively remove unwanted signals or maximize SNR depends on the association of the channel coefficient matrix EL describing the channel response between the transmitting antenna and the receiving antenna. Sexual channel, time-spaced interference in the channel causes inter-symbol interference (IS I). To improve performance, a receiver unit attempting to reply to a particular transmitted data stream will need to improve "interactive interference" from other transmitted signals, and Both inter-symbol interference from other transmitted signals, etc. To handle crosstalk and intersymbol interference, Space processing (which can properly handle interactive interference but can't effectively deal with inter-symbol interference) can be replaced by space time processing -39- 1269549 _ (35) I send a cigarette. In the specific example, MMSE linear, etc. The MMSE-LE can be used for spatial time processing of a scatter channel. With MMSE-LE technology, spatial time processing assumes a similar pattern to the spatial processing of non-scattering channels. However, as will be further explained below, space Each "filtering specification" packet in the processor is not only specified. When the channel estimation value (ie channel coefficient matrix) is accurate, it is more efficient to use MMSE-LE technology for spatial time processing. In another embodiment A decision feedback equalizer (DFE) can be used for the spatial time processing. The DFE is a nonlinear equalizer that has functions for channels of severe amplitude distortion and uses decision feedback to eliminate the detected symbols. Interference between symbols if the data stream can be decoded without error (or with minimal error)' then the intersymbol interference generated by the modulation symbol corresponding to the decoded data bit can be valid In a further embodiment, a maximum likelihood sequence estimator (MLSE) can be used for the spatial time processing. When the channel estimates are not so accurate, the DFE and MLSE techniques can reduce or possibly eliminate performance degradation. The DFE and MLSE technologies will be further elaborated in the literature by SL Ariyavistakul et al., entitled "Optimal Space Time Processor with Dispersive Interference: Integrated Analysis and Demand Filter Development", July 1999 IEEE Communications Monthly Volume 7 and incorporated herein by reference. Figure 4A is a block diagram of an RX ΜΙΜΟ/data processor 260a - a particular embodiment of a base station 104 or terminal 106 of Figures 2A and 2B A specific embodiment of the serving. The signals transmitted from the (up to) Ντ transmit antennas are received by the N r antennas 252a through 25 2r , and then the paths are assigned to the respective 1269549 rounds. The demodulator page (36) of the demodulator 2 5 4 (also Called the front-end processor). Each demodulator 254 adjusts (e. g., filters and amplifies) the respective received signals, each downconverts the conditioned signal to a relay frequency or fundamental frequency, and then digitizes the down converted signal to provide a data sample. Each demodulator 254 may further transform the data samples with a return pilot to produce a received modulated symbol stream that will be provided to a spatial/spatial time processor 410.

If OFDM is used for data transmission, each demodulator 254 further implements a process complementary to that implemented by the modulator 2 22 shown in Figure 3E. In this case, each demodulator 254 includes an FFT processor (not shown) that produces a conversion indication of the data samples and provides a stream of modulated symbol vectors. Each vector includes an NF modulation symbol for the NF frequency subchannel, and a vector is provided for each time slot. The modulated symbol vector streams from the FFT processors of all NR demodulators are provided to a demultiplexer/combiner (not shown in Figure 4A), which is first "channelized" from each FFT processor. The modulated symbol vector stream becomes the majority (up to NF) modulated symbol stream. For a transport processing scheme in which each frequency subchannel is independently processed, the demultiplexer/combiner provides each (up to) Nf modulated symbol stream to a respective spatial/spatial time processor 410. For a system that does not use OFDM, the spatial/spatial time processor 410 can be used to perform ΜΙΜΟ processing on the modulated symbols from the NR receive antenna. For a system using OFDM, the spatial/spatial time processor 410 can be used to perform ΜΙΜΟ processing on the modulated symbols from the NR receive antennas for data transmission for each NF frequency subchannel. Alternatively, the spatial/spatial time processor 410 can be used to perform ΜΙΜΟ processing on the modulation symbols for all NF frequency subchannels, for example, at a time of -41 - 1269549 (37). Dependent <: ::3⁄43⁄4毅:; hair:; Xiang Yiyi:? Yi Yi $;: multiplex processing. 1. CCM1 technology (spatial processing) In a system with a Ντ transmit antenna and an NR receive antenna, the received signal at the output of the NR receive antenna can be expressed as: r - Hx + η Equation (6)

Where L is the received symbol vector (ie, the NR χ 1 vector of the output channel is measured at the receiving antenna), and the NR X Ντ channel coefficient matrix representing the response of the τ transmitting antenna to the NR receiving antenna channel at a specific time , & is a symbol vector (ie, Ντ X 1 vector input to the channel), and η represents an NR X 1 vector of noise plus interference. The received symbol vector L includes an NR modulation symbol within the number received by the receiving antenna at a particular time slot. Similarly, the transmitted symbol vector 2L includes the Ντ modulation symbol in the τ signal transmitted by the Ντ transmitting antenna at a particular time slot. For CCMI spatial processing techniques, the receiver unit first performs a channel matching overrun operation on the received symbol vector L. The filtered output can be expressed as: HHr - HhHx + HHn Equation (7) where the superscript " Η π represents the shift term and the conjugate complex number. A square matrix I can be used to represent the product of the channel coefficient matrix Η and its conjugate shift term KH ( That is, K = iLHiD. The channel coefficient matrix £L can be transmitted, for example, by transmitting the preamble along with the AC data. To implement the "best" reception and estimation of the SNR of the transmission channel, it is usually customary to insert a known preamble (eg, all orders). The preamble is transmitted to and from the one or more transmission channels. The method of estimating a single transmission channel based on a pilot signal and/or data transmission can be used in many of the prior art techniques - 42 - 1269549 (38) It is known in the literature. One of the channel estimation conveniences is described by F. Ling in the title "Best Reception, Performance Range and Cutoff Rate Analysis Reference Auxiliary Symbiosis CDMA Communication and Application", IEEE Communications Monthly, October 1999 issue. Or other channel estimation techniques can be derived as a matrix type to derive the channel coefficient matrix ϋ.

The transmission symbol vector (I) estimate can be obtained by multiplying the matched filter vector ilH L by the reciprocal (or imaginary-reciprocal) of the square matrix Ε_, which can be expressed as:

i - R-1HHL = 2L + Equation (8) = x. + iLf It can be observed from the above equation that the transmitted symbol vector 1 can be recovered, and the symbol vector L is received by matching filtering (that is, multiplied by the matrix ilH first) and then filtered. The knot is first multiplied by the reciprocal of the square matrix 1.

The SNR of the transmission channel can be determined as follows: The auto-associated matrix inn of a noise vector il is first calculated from the received signal. Usually, inn is a Hermitian matrix, that is, its symmetric conjugate complex number. If the components of the channel noise are uncorrelated and further independent and equally distributed (iid), the automatic correlation matrix Inn of the noise vector il can be expressed as:

Inn = , and Equation (9) tnn 〇; where U is the identity matrix (ie, 1 along the diagonal, otherwise 0) and σ = the amount of noise variation in the received signal. The auto-correlation matrix ln,n, -43 - 1269549 (39) of the post-processing noise vector iL' (ie after matched filtering and pre-multiplied by the matrix BJ1) can be expressed as:

In n =E[|L'iL'H] Is the performance of the buyer program (10) From the equation (1 0), the noise variation σ of the i-th component of the post-processing noise iL' is equal to σ丨r, where r is the i-th diagonal line element of 1. For a system that does not use a a OFDM, the i-th element represents the ith receive antenna. If OFDM is used, the subscript "i" can be decomposed into subscript "jk", where "j" represents the jth space subchannel and "k" represents the kth frequency subchannel. For CCMI technology, the SNR of the ith element of the received symbol after processing (ie, the i-th component of i) can be expressed as:

2 Equation (11) If the variation of the ith transmission symbol |4_|2 is equal to (1.0), the SNR of the received symbol vector can be expressed as: The amount of noise variation can be 1/^ normalized to receive the symbol vector i component. If a modulated symbol stream is replicated and transmitted through multiple transmit antennas, then these modulated symbols can be added together to form a combined modulation symbol. For example, if a data stream is transmitted via all antennas, the modulation symbols corresponding to all ττ transmit antennas will be summed, and the combined modulation symbol can be expressed as: Equation (12) ί=*1 rii -44- 1269549 (40) I buy and buy

Alternatively, the transmitter unit can be operative to transmit one or more data streams over a plurality of transmission channels or to use the same coding and modulation scheme on some or all of the transmission antennas. In this case, only one S N R (e.g., average SNR) is required for all transmission channels. For example, if the same coding and modulation scheme is applied to all transmit antennas (e. g., using selective channel reversal), the SNR of the combined modulation symbol SNRtQtal can be derived. This SNRtQtal will then have a maximum combined SNR equal to the sum of the SNRs of the signals from the NR receive antenna: the combined SNR can be expressed as:

,=Λς:7 Equation (13) ση /-1

Figure 4B is a block diagram of a spatial/spatial time processor 4 10b, which is capable of implementing CCMI techniques. Within the spatial/spatial time processor 410b, the received modulated symbol vector 1 stream from the NR receive antenna will be provided to and filtered by a match filter 412 which pre-multiplied each vector l by its conjugate-shift term channel The coefficient matrix iiH is as shown in equation (7). The filtered vector is further pre-multiplied by a multiplier of a square matrix gj1 by a multiplier 4 1 4 to form an estimated value i of the transmitted modulation symbol vector 2L, as shown in equation (8) above. The estimated modulation symbol oracle is provided to a predictable channel coefficient matrix Η (for example, based on a pilot signal similar to a conventional pilot-assisted single and multiple carrier subsystem, as is known in the art). 1 8. In general, the channel coefficient matrix 预估 can be estimated based on the preamble or communication data or the modulation symbols of both. The channel coefficient matrix Η is then provided to a matrix processor 420 which can derive a square matrix R as described above. -45 - 1269549

(41) The far-predicted modulation symbol J and/or the combined modulation symbol representation will also be provided to a CSI processor 448' which determines whether the full or partial CSI will be used for the transmission channel. For example, the 'CSI processor 448 can estimate the SNR of the ith transmission coherence matrix 虻η according to the received preamble signal and then calculate the SNR. The SNR of the transmission channel contains a portion of the CSI that can be reported back to the transmitter unit. For some transport processing schemes, the symbol streams from all or most of the antennas used to transmit a data stream may be provided to combiner 4i6, which combines redundant information across time, space and frequency. The combined modulation symbol is then provided to the RX data processor 480. For other modes of communication, the estimated modulation symbol i can be provided directly to the RX data processor 480 (not shown in Figure 4B).

The spatial/spatial time processor 410 thus produces one or more independent symbol streams corresponding to one or more transport streams. Each symbol stream includes a post-processed modulation symbol whose transmitter unit corresponds to the modulation symbol and is estimated for it before processing the full/partial C S I . This (post-processed) symbol stream is then provided to RX data processor 480.

FIG. 4A shows a specific embodiment of an RX data processor 480. In this particular embodiment, a selector 482 receives one or more symbol streams from the spatial/spatial time processor 410 and extracts the modulation symbols corresponding to the demand data stream to be replied. In another embodiment, the RX data processor 480 has a modulated symbol stream corresponding to the data stream of the demand, and the modulated symbol extraction is performed by the combiner 4 1 in the spatial/spatial time processor 410. 6 and implemented. In any event, the extracted modulated symbol stream will be provided to demodulation element 484. For a specific embodiment in which the data stream of each transmission channel is encoded and modulated independently (for example, according to the S N R of the channel), the response of the selected transmission channel is adjusted to -46 - 1269549. __ (42)

The variable symbols are demodulated according to a demodulation variant (e.g., M-PSK, M-QAM) complementary to the modulation scheme used for the transmission channel. The demodulation data from demodulation element 484 is then deinterleaved by a deinterleaver 486 in a manner complementary to the implementation of the transmitter unit, and the deinterleaved data is further passed by a decoder 488. Decoding is performed in a complementary manner to the implementation of the transmitter unit. For example, an acceleration decoder or a Viterbi decoder can be used as the decoder 48 8 if the accelerometer or whirling code is implemented at the transmitter unit, respectively. The decoded data stream from decoder 48 8 represents an estimate of the transmitted data stream being replied to. 2. MMSE technique (spatial processing) For MMSE techniques, the receiver unit implements an initial MMSE estimate of a received symbol vector l pre-multiplied by a matrix to derive the transmitted symbol vector 2L, expressed as: X = Mr equation (14 )

The matrix is selected such that the mean squared error between the initial MMSE estimate i and the transmitted symbol vector 2L (ie, an error vector L between the two is minimal. The matrix Μ can be expressed as: Μ = ΗΗ(ΗΗΗ + inn)·1 Equation (15) According to equations (14) and (15), the initial MMSE estimate X of the transmitted symbol vector 2L can be determined as: Equation (16) X == Mr -Ηη(ΗΗη + φηη)-1Γ -47-

1269549 (43) I does not deviate from the minimum mean square error estimate (magic, can be obtained by pre-multiplying the initial estimate X by a pair of angular matrices D; 1), as follows: X - D : 1! , equation (17) \ where

Dv—^ diag(l/vu, l/v22,...,1/vNrN) and the diagonal element of the vii matrix, which can be expressed as: V - Hh + Η^Η)-1 Received symbol vector after processing (i, the i-th element of i) can be expressed as: SNR/=feil Equation (18) uu where Un is the variation of the i-th element of the error vector S, which is defined as e = 2L* £ ? and the matrix LL Can be expressed as:

Liao = i - 1 Equation (19) If the average of the variation (@) of the ith transmitted symbol Xi is equal to 1 (1.0) and the equation (19) is = 丄-1, the SNR of the received symbol vector can be expressed as : SNRi=-^- Equation (20) 1 - The estimated modulation symbols (g) can be combined to obtain a combined modulation symbol left, as described above for the CCM1 technique. -48 - 1269549

(44) Figure 4C shows a spatial/spatial time processor 4 1 0 c - a specific embodiment that is capable of implementing MMSE techniques. Similar to CCMI technology, the matrix In and Inn can be estimated first based on the received pilot signal and/or data transmission. The weighted coefficient matrix win will be calculated according to equation (15). In the spatial/spatial time processor 410c, the received modulated symbol vector L stream from the NR receive antenna is pre-multiplied by a multiplier 422.

To form an initial estimate of the transmitted symbol vector, to the above equation (I6). The initial predicted stem system is further pre-multiplied by a scale matrix by a multiplier 424 to form a non-offset estimate of the transmitted symbol vector £, as shown in equation (17) above.

Again, for some transmission processing schemes, a plurality of streams corresponding to the transmitted modulation (the estimated modulation symbol i stream of many transmit antennas can be provided to a combiner 4 2 6, the combination spans time, space and frequency Redundant information. The combined modulation symbol will then be provided to the rX data processor 48. For some other transmission processing schemes, the estimated modulation symbol | may be provided directly to the RX data processor 480. The RX data processor 480 demodulates, deinterleaves, and decodes the modulated symbol stream corresponding to the recovered data stream, as described above. The estimated modulation symbol i and/or the combined modulation symbol i will also be provided to the CSI processing. The 448' determines whether the modulo or partial csi is used for the transmission channel. For example, the CSI processor 448 can estimate the SNR of the ith transmitted signal according to equations (18) through (20). The SNR for transmitting the signal includes a return to A portion of the CSI 〇 estimated modulation symbol i of the transmitter unit is further provided to an adaptive processor 42 8 which can derive matrix Μ and oblique line matrix -49 according to equations (15) and (17), respectively. - 1269549 (45)

3. mmse-le technology (space time processing, multiple inter-time processing techniques can be used to process signals received over a time-scattered channel. These techniques include the use of time domain channel equalization techniques such as MMSE-LE, DFE, The MLSE and other possible techniques incorporate the above-described spatial processing techniques for non-dispersive overnight. The spatial time processing is implemented on the Nr input signal at the RX ΜΙΜΟ/data sink 260. When time scatter occurs, the channel The coefficient matrix 氐 can be obtained in a delay range, and the elements of the matrix 作为 are used as a linear conversion function rather than a coefficient. In this case, the channel coefficient matrix Η can be written as a channel conversion function matrix ϋ(τ). It can be expressed as: H(1) is called hg(T)} for 1 si SNR, and 1 SNt is the equation (21) whose center (1) is a linear transfer function from the jth transmitting antenna to the ith receiving antenna. As a result of (1), the received signal vector £(the system channel transfer function matrix ϋ(τ) and the transmitted signal vector χ(1) are convoluted, which can be expressed as ··r(t) = \Urmt-r)dr Equation (22) as a solution Tune A portion of the variable function (implemented by demodulator 254 in Figure 4A), the received signal is sampled to provide a received sample. Without loss of versatility, the time-scattering channel and the received signal can be described below. It is expressed in a discrete time representation. First, the channel transfer function vector b(k) of the transmission antenna at delay k can be expressed as: -50 - 1269549 (46) PP continuation page kj(k) = [hdk H2j(k) Λ h~(k)]T is used for OSkSL equation (23) where h^k) is the k-th tap weight of the channel transfer function, and the path between the jth transmission line and the ith receiving antenna And the maximum range of L system channel time diversity (within the sample interval). Secondly, the NRxNT channel transfer function with delay k, the matrix can be expressed as: ii(k) = [kKk) 3b(k) Λ h ~(k)] for OSkSL equation (24) Receive signal vector t at sample time η (n) can be expressed as: · r(n) = 2]H(A:)x(/7-k) + n(n) = Ηχ(τ?) + n(n) Equation (25) k=0 Where g is an NR x (L+1)NT block configuration matrix representing the channel matrix transfer function of the sampled sample (k), and can be expressed as: g = (9) S(1)AH(L)] (n) The order of the L+ 1 vector of the received sample captured in the L+ 1 sample interval

Column, where each vector contains N r samples relative to the N r receive antenna, and can be » Table TF is · l(n) do-1) Μ _x(n - L) An initial estimate of the transmitted symbol vector at time η The value to (η) can be derived by implementing a convolution of the sequence of the received signal vector L(n) with the 2Κ+1 sequence and the NRX Ν τ weighting matrix MX k), as follows: -51 - 1269549 (47)发?第_•绩·Κ X(九)=(九)咖-幻 = Equation (26) where this = [ΜΧ-Κ) eight M(o) Λ Μ(κ)], the system determines the equalizer filter delay range Parameter, and >(η + Κ) Μ一r(«)= L(n) Μ U(卜Κ)

The sequence of weighting matrices Kk) is selected to minimize the mean square error. The solution of MMSE can be illustrated as a sequence of weighting matrices M_(k) that satisfies the following linear limits: Female =-κ 〇, Η^(-λ), 〇, - Κ$λ< - L -L<X<0 0&lt ;λ<Κ Equation (27) where K(k) is a sequence of NR χ NR spatial time correlation matrices, which can be expressed as: R(k) = E{r(n - k)rH (η)} = ^

Min( L,L—A) m-riwx(0,,*) -L^k<L Equation (28) Other cpnn(k) is a noise automatic correlation function, which can be expressed as: ^n(k) = Ε{η(λ-^)ηΗ(λ)} Equation (29)' For white (temporarily uncorrelated) noise, ln(k) = ιηδ(1〇, where in this case only the spatial correlation matrix is represented For spatially and temporarily unassociated noise and the power at each receiving antenna is equal, ^(k)= -52- 1269549 (48) Code continuation page·· σ2[δ(1〇. Equation (27) can be further expressed as: 昼' or heavy"a;1 Equation (30) where E = is the Toeplitz square and the square j, k is given by K(j - k), And H(L)

S(L,1)

Μ H(〇)

Qk.NrxNt where Lmxn is a mxn zero matrix. The non-offset MMSE-LE estimate i(n) of the transmitted signal vector at time n can be expressed as: x(«) = DvX(^) = DvMr(^) Equation (3 1 )

D;1 =diag(\/vu ,1/ν22 ?Λ,1/νΝτΝτ) Equation (32) where νυ 矩阵 matrix V ith diagonal element 〇 “a scalar quantity”, which can be expressed as: V- MH^R Η Equation (33) The error covariance matrix can be expressed as follows: = = [ί(8)一5ν Mr(zz)][x(/z)-r^^M^Dy ]} Equation (34) = I-D"vY-VDv +DyVDy -53 - 1269549 (49) The estimated S Ν R for the symbol transmitted on the ith transmit antenna can be expressed as: Equation (35) MMSE- The LE technique can be implemented by the spatial/spatial time processor 410c in Figure 4C. In this case, the multiplier 4 2 2 implements a sequence of received signal vectors L(n) and a sequence of weighting matrices M_(k) Cyclotron to obtain an initial estimate i(n) as shown in equation (26). Multiplier 424 implements initial estimation to (n) pre-multiplied by a diagonal matrix to obtain a non-offset mmSE-LE prediction box (η) 'As shown in equation (31). The adaptive processor 428 derives the weighting matrix rainbow (k) shown in equation (30), and the I diagonal matrix shown in equation (32). The processing can be based on the above MMSE technology and the like. In this way, the Snr of the symbol stream transmitted by the first transmitting antenna can be estimated by the CSI processor 448 according to equation (35). 4. DFE technology (space time shift j Figure 4D space/space time processor 41〇 d is a block diagram of a specific embodiment which is capable of applying DFE technology to the male. In the spatial/spatial time processor 41〇d, the received modulation symbol vector L(n) stream from the & receiving antenna is positive Processing is processed by the receive processor 423 to provide predictive modulation symbols for the data stream to be replied. The forward receive processor 432 can implement the CCMI or MMSE techniques described above or some other linear space equalization technique. 434 then combines the estimated distortion components provided by a feedback processor 44 and the estimated modulation symbols from the forward receive processor 43 2 to provide an "equivalent" modulation symbol that substantially removes the distortion components. Initially, the estimated loss -54 - 1269549 hair 瞵 黉 黉 ( (50) The true component is zero and the equalization modulation symbol is only equal to the estimated modulation symbol. The equalization modulation symbol from the adder 4 3 4 Estimated value of the transmitted symbol vector 2L X 〇

For some transport processing modes, a majority of the estimated modulated symbol i streams corresponding to the majority of the transmit antennas used to transmit a data stream may be provided to a combiner 433, the combination of which spans time, space and frequency. Redundant information. The combined modulation symbol is then provided to the RX data processor 4800. For some other transmission processing methods, the estimated modulation symbol i can be directly supplied to the RX data processor 480. The RX data processor 480 demodulates, deinterleaves, and decodes the modulated symbol stream with respect to the data stream that will be replied, as illustrated in Figure 4A above.

The decoded data stream is also encoded and re-modulated by the channel data processor 433 to "remodulate" the symbol, which is the estimated value of the modulation symbol at the transmitter. The channel data processor 438 performs the same processing (e.g., encoding, interleaving, and modulation) on the data stream as at the transmitter. The remodulated symbols from the channel data processor 43 8 are provided to a feedback processor 440 which processes the symbols to derive the estimated distortion components. The feedback processor 440 can implement a linear space equalizer (e.g., a linear cut equalizer). For DFE technology, the decoded data stream is used to derive an estimate of the distortion produced by the decoded information bits. If the data stream is decoded without error (or with minimal error), the distortion component can be accurately predicted, and the intersymbol interference generated by the decoded information bits can be effectively eliminated. The processing performed by the forward receive processor 43 2 and the feedback processor 440 is usually adjusted at the same time to minimize the mean square error of the intersymbol interference within the equalized modulation symbol - 55 - 1269549 _ - (51) I Xia Zhu page difference (MSE). The DFE treatment is further described in detail in the aforementioned literature by Ariy avistakul et al. For the DFE technique, an initial estimate of the transmitted symbol at time η, i(n) vector, can be expressed as: X(8) = + - exempt) Equation (36) where L(n) is the receive modulation in equation (25) above The symbol vector, χ(η) is provided by the RX data processor 480, and the symbol decision vector, Mf(k), S k S 0 is used by the forward receiving processor 43 2 (Κβΐ) (NT x NR) pre-coefficient The sequence of matrices, and Mb(k), 1 S k S K2, is a sequence of inverse 2-(>^\1^) feedback coefficient matrices used by feedback processor 440. Equation (36) can also be expressed as: Equation (37)

x(n) = r(n) + \{n)

Μ / =咝(-κ,) Μ(-K, +1)八M(0)],Μ广[Μ(1) Μ(2) λ M(K2)], do-1) _ "Κη + Κ, )" ί(π-2) , and £(spoon = rO + Ki-i) Μ Μ step - κ2)_ _ r(8) , if the MMSE standard is used to find the coefficient matrix, the mean square error It, and M are minimized The answer to =b is ready to use. The MMSE solution for pre-filtering can then be expressed as: II I"S_1 Equation (38) -56 - 1269549 _ (52) I 瞵 _ 续 Η Η: Q(Kt-L)NaxMT H(L) H (Ll) MH(0) and I is a (Ki + 1) NR X (Ki + 1) NR matrix composed of NR X NR blocks. The first (i,j) block in 昼 is defined as: κ,-/+ι Ι(〇·)= Σ Η(10)+ /-7Ησ2ί-/) Equation (39)

The MMSE solution for the feedback filter: μ#)=- ΣΑωιι(卜力, i<k<K2y-κ, equation (40) MMSE-LE as described above, one without deviation prediction | (n) can be expressed as : X(n) = Dvdfc x(n) ~ + Equation (41) where iVdfc ·

Wag (Vdfc l!, V吮, 22, Λ, Vdfc, "r/Vr) Equation (42) and VdtwHI, Y_dfe ith diagonal element, which can be expressed as Equation (43) ~ ~ Μ \ ~

Yafc=MrH = H R Η The composite error covariance matrix is defined as: -57- 1269549

(53) = Ϊ -SvarcXafc - VdfcD-ldfc, and bucket m, 10,000 program (44) The estimated SNR for the symbols transmitted on the ith transmit antenna can be expressed as: SNR- —= --^ Equation (45 )

Wdlb, 7 1-Vdfcj/

5. Complete CSI technique (spatial processing) For the complete CSI technique, the received signal at the output of the receiver antenna can be expressed in equation (6), which is: r = Hx + η from the channel matrix and its conjugate shift term The eigenvector decomposition of the Hermitian matrix formed by the product can be expressed as: Ηη Η = ΕΛΕη Equation (46)

Among them, the eigenvector vector matrix and the pair of angular matrices of the eigenvalues are both Ντ X Ντ. The transmitter uses the eigenvector matrix E to pre-set a set of Ντ modulation symbols k as shown in equation (5) above. The pre-adjusted modulation symbol transmitted by the Ντ transmission antenna can thus be expressed as: X = Equation (47) Since 2^ over _ system Hamilton, the eigenvector matrix is a unit matrix. Therefore, if the elements of k have equal powers, the elements of 2L have equal powers. -58- 1269549 玛玛锶秘页 (54) The received signal can be expressed as: r - HEb + a Equation (48) The receiver implements a channel-match-filter operation, which is then pre-multiplied with the correct eigenvector. The receiver implements a channel-match-filter and pre-multiplied operation result as a vector 1, which can be expressed as: Equation (49) z - EH HH HEb + EH ΗH n =AbL + n1

The covariance of the new noise term can be expressed as: Equation (50) Ε{ήήΗ) = E(Eh Hh nnH HE) = ΕΗ Ηη HE = Λ That is, the noise component is not related to the variation given by the eigenvalue. The SNR of the i-th component of I is λ, which is the i-th diagonal element of i. The complete CSI processing system is further described in detail in U.S. Patent Application Serial No. 09/532,492

The spatial/spatial time processor embodiment shown in Figure 4B can be used to implement the full C S I technique. The received modulation symbol vector is filtered by a matching filter 4 1 2 which multiplies each vector [first multiplied by the conjugate shift term channel coefficient matrix ϋΗ as shown in equation (49) above. The filter vector is further multiplied by the multiplier 4 14 to the correct feature vector to form an estimate of the modulation symbol vector L, as shown in equation (49) above. For the full CSI technique, the matrix processor 420 can be configured to provide the correct feature vector. Subsequent processing (e.g., by combiner 416 and RX data processor 480) can be implemented as described above. For the complete C S I technology, the transmitter unit can be given according to the characteristic value -59- (55) 1269549

The S N R selects a coding scheme and a modulation scheme (i.e., a signal group) for each feature vector. The constraint is that the channel condition does not change the time between the f and the reporting at the receiver unit, and the interval between the time it is used to pre-condition the transmission at the transmitter unit, the performance of the communication system Can be equal to a set of independent AWGN channels with known SNR. 6. Receiver Processing For the hold, $/branch receiver processing technique, the initial NR receive signal is processed to continually reply a transmit signal at ~ time. As each transmitted signal passes back k 〃 will be removed (ie, eliminated) by the received signal before the next transmitted signal is returned. If the transmitted data stream can be decoded error-free (or with minimal error) and if the overnight response is estimated Reasonably accurate, it can effectively eliminate the interference of the receiving L 5 tiger due to the previously transmitted transmission signal, and the SNR of each transmitted signal that will be recovered by the performance will be improved. In this way, all transmissions can achieve higher performance (possibly except for the first transmission signal that will be replied to). Figure 5 is a flow chart of the keeper who does not continuously eliminate the receiver processing technique to process the NR received signal to respond to Ν τ transmission as * station a %, τ 疋仏唬. For the sake of simplicity, the following figure 5

It is assumed that (1) the number of transmission channels is equal to the number of transmitting antennas (i.e., NC = NT ' and the transmission channel is a spatial sub-channel of a system that does not use OFDM)' and (2) each transmitting antenna transmits an independent data stream. Initially at step 512, the receiver unit performs spatial and/or spatial time processing on the NR received signal intended to separate the multiple transmitted signals included in the received signal. Spatial processing can be performed on the received signal if the channel is non-dispersive. It may be required or required to receive signals on the -60-1269549 fascination (56) to implement linear or nonlinear transient processing (ie, spatial time processing) if the ΜΙΜΟ channel is time-diversity. This spatial processing can be based on CCMI, MMSE or some other technology, while spatial time processing can be based on MMSE-LE, DFE, MLSE or some other technology. The number of signal isolations that can be achieved depends on the number of associations between the transmitted signals, and if there is less correlation between the transmitted signals, there can be more signal isolation.

The spatial or spatial time processing provides a Ντ "post-processing" signal, which is an estimate of the τ τ transmitted signal. The SNR of the processed signal after step 5 14, τ will be determined. In one embodiment, the 16 SNR is ranked sequentially from highest to lowest SNR in step 5, and the processed signal is selected and further processed (i.e., "detected") to obtain a decoded data stream with the highest SNR. This detection typically involves demodulating, deinterleaving, and decoding the selected processed signal. The decoded data stream is an estimate of the data stream on the transmitted signal replied to by this iterative process. The particular post-processing signal to be detected may also be selected in accordance with some other scheme (e.g., by a schedule or transmitter unit). In step 5 18, whether all transmitted signals have been replied will be determined here. If all transmitted signals have been replied, the receiver processing will terminate. Otherwise, the interference due to the decoded data stream will be estimated and removed by the received signal to produce a "modified" signal for the next iteration to reply to the next transmitted signal. In step 520, the decoded data stream is used to form an estimate of the interference due to the transmitted signal on each received signal (with respect to the data stream just decoded). The interference can be estimated by first re-encoding the decoded data stream, handing over -61 - 1269549, saying _ continued (57)

The re-encoded data and symbols map the interleaved data (this data stream uses the same decoding, interleaving, and modulation methods as used at the transmitter unit) to obtain a "remodulated" symbol stream. The remodulated symbol stream is an estimate of the modulated symbol stream previously received by the NR receive antenna and transmitted by the NR receive antenna. The remodulated symbol stream is then derived from each of the N R elements within the channel response vector I to derive an interference signal due to the jth recovered transmission signal. At step 522, the vector kj is a particular row of the (NR X Ντ) channel coefficient matrix Η. The > ^ interference signal is then subtracted from the off received signal to derive the NR modified signal. These modified signals represent the signals at the receiving antenna if they are not transmitted due to the components of the decoded data stream (i.e., it is assumed that interference cancellation is effectively implemented). The processing at steps 512 through 5 16 will then repeat (rather than receive) the NR modification signal to reply to the other transmission signal. Steps 5 12 through 5 16 will therefore repeat for each of the transmitted signals to be recovered, and steps 520 and 5 22 will be implemented if another transmitted signal is to be recovered.

For the first iteration, the input signal is derived from the Nr receive signal of the Nr receive antenna. For each of the iterations, the input signal is derived from the previous iteration of the signal from the interference canceller. The processing of each iteration is performed in a similar manner as the input signal is appropriately replaced. More specifically, in the case of each iteration after the first iteration, the signal hypothesis in the previous iterative detection will be eliminated, so the range of the matrix matrix of the subsequent iterations will be reduced. The receiver processing technique is continuously eliminated and thus multiple iterations are performed, each of which will repeat the transmitted signal one time. Each iteration (except the last) is implemented in one or two parts. -62 - 1269549 Sending an actor says that the product is processed to recover a signal and the modification signal required to generate the next iteration. In the first part, spatial processing or spatial time processing is performed on the NR received signal to provide a τ post-processing signal, and one of the post-processed signals is detected to recover the data stream corresponding to the transmitted signal. In the second part (which will not need to be implemented in the last iteration), the interference due to the decoded data stream will be removed from the received signal to derive a modified signal that the reply component has been removed. Initially, the first iteration of input signal L1 is a received signal L, which can be expressed as: η Μ Equation (51) These input signals are linear or linearly processed to provide a τ post-processed signal I1, which can be expressed as: fm I Txl 13⁄4 Μ l^r ______” Equation (52) The SNR of the post-processed signal can be estimated, which can be expressed as: Equation (53) One of the post-processed signals is selected for further processing (eg, processing the signal with the highest SNR) ) to provide a decoded data stream. The decoded data stream will then be used to estimate the interference i1 generated by the reply signal, which can be expressed as: -63 - 1269549 (59) 1 " Μ Equation (5 4) is defined in this iteration, interference i1 It will be subtracted from the input signal vector Li to derive a modified signal containing the input signal vector L2 for the next iteration. Interference cancellation can be expressed as:

Equation (55)

The same process will then be repeated in the next iteration, and the vector contains the input signal for this iteration. In order to continuously eliminate the receiver processing scheme, each iteration will reply a transmission signal, and the SNR of the jth transmission signal will be recovered in the kth iteration, and < can be provided as the transmission channel used for this reply signal. CSI. For example, 'If the first post-processing signal X丨 is replied in the first iteration, the second post-processing signal X is replied in the second iteration, etc., and the % post-processing signal V will be at the end. In the case of replies, the CSI that can be reported for these reply signals can be expressed as: r = 〇;,,22,., to]. 4E is a block diagram of an RX®/data processor 260e capable of implementing a continuous cancellation receiver processing technique. The signals transmitted by the (up to) 传送τ transmit antennas are received by the NR antennas 252a through 252r and assigned paths to respective demodulators 254. Each demodulator 254 processes the respective received signals and provides a received modulated symbol stream to the RX ΜΙΜΟ/data processor 26〇e. For systems using OFDM, RX MIM〇/Sco processor 260e -64 - 1269549

(60) may be used to process the NR modulated symbol stream from a receiving antenna for each NF frequency sub-pass for data transmission. For systems that do not use 〇FDM, the system 'RX ΜΙΜΟ/data processor 26〇e can be used to process the NR modulated symbol stream from the Nr receive: antenna. < In a specific embodiment not shown in Figure 4E, the mM®/data processor 260e includes a plurality of continuous (i.e., in series) receiver processing stages 45, each of which transmits a phase for recovery. In a transport processing scheme, data streams are transmitted on each transport channel, and each data stream is processed independently (e.g., using its own coding and modulation scheme) and transmitted by its respective transport antenna. For this transport processing scheme, the number of data streams is equal to the number of transmitted signals 'is also equal to the number of transmit antennas for data transmission (which may be available (a subset of transmit antennas). To clearly indicate this transport processing scheme, RX The data processing unit 260e is hereby described. Each receiver processing stage 450 (except the final stage 45〇n) includes a channel/data processor 460 coupled to an interference canceller 47A, and the final stage 450n includes only channels/ The data processor 46 〇 η. For the first receive modulator processing stage 450a 'channel ΜΙΜΟ / data processor 46 〇 a receives and processes the NR modulated symbol stream from the demodulators 254a through 25 to provide the first transfer 7 [Lü No. 1 decodes the data stream. For each of the second to final stages 450b to 450n, the channel/data processor 46 of this stage receives and processes: the modified symbol of the interference canceller from the previous stage Streaming, to derive: a decoded data stream for replying to the transmitted signals by the stage. Each channel/data processor 460 further provides a csi (eg, SNR) for the associated transmission channel -65 - (61) 1269549

For the first, a receiver processing stage 450a, the interference canceller 470a receives the NR modulated symbol streams from all of the modulators 254s. For the second to last I, the 'interference canceller 4' receives the symbol stream from the previous stage of the interference cancellation 'R, and the second modification. Each interference canceller 470 receives the decoded data stream from the phase, 'Θ期, s_*, u, 遒MIM0/data processor 460, and performs processing (storage, 1 4 such as encoding, interleaving, modulation) , channel response, etc.) to derive the Y system ^1U ^ ^ ^ < estimated nr remodulation symbol stream, since the receive modulator 2 corresponds to the decoded data stream. The modulating symbol stream will then be subtracted from the received modulo _##: μ to derive a stream of Nr modified symbols including all of the subtracted (i.e., eliminated) < interference components. The Nr modified symbol stream will then be provided to the next stage. A controller 27 shown in Figure 4 is coupled to the RX μίμο/data processor 26〇e and can be used to dominate the various steps within the receiver cancellation process. Figure 4E shows that each data stream is transmitted through its own transmit antenna and can be used in a direct receiver configuration (i.e., a stream corresponding to each transmitted signal). In this case, each receiver processing stage 45 0 can be operated to reply to one of the transmitted signals and to provide a decoded stream corresponding to the reply transmitted signal. For some other transport processing schemes, a data stream can be transmitted through multiple transmit antennas, frequency sub-compasses, and/or time intervals to provide spatial, frequency, and time diversity, respectively. For these schemes, the receiver process begins by deriving a receive modulation stream of signals transmitted on each of the transmit and receive antennas. For multiple transmit antennas, frequency subchannels and/or time zones -66 - 1269549

The modulation symbols between (62) can be combined in a manner complementary to the multiplexed implementation at the transmitter unit. The stream of combined modulation symbols is then processed to provide a relatively decoded data stream. Figure 4F is a block diagram of one embodiment of a channel ΜΙΜΟ / data processor 460x that can be used for each of the channel ΜΙΜΟ / data processors 460a through 460n in Figure 4E. In this particular embodiment, processor 46A includes a spatial/spatial time processor 410 coupled to RX data processor 480.

The spatial/spatial time processor 41 performs spatial or spatial time processing on the Nr input signal. The spatial/spatial time processor 410 can implement CCMI, MMSE, or some other spatial processing technique for non-distributive channels, and can implement MMSE-LE, DFE, MLSE, or some other spatial time processing technique for a non-distributive channel. . Figure 4G is a block diagram of an interference canceller 470x, which may be used with each of the interference cancellers 470 of Figure 4E. In the interference canceller 47A, the decoded data stream (k) from the channel/data processor 46 in the same phase is re-encoded, interleaved and re-modulated via a data processor 210 to provide retuning. Variable symbol, which is an estimate of the modulation symbol at the transmitter before MIM〇 processing and channel distortion. The τ χ data processor 21 〇乂 implements the same processing (e.g., decoding, interleaving, and modulation) of the data stream at the transmitter unit. The remodulation symbol is then provided to a channel simulator 4 7 2 ' which processes the symbol in an estimated channel response to provide an estimate of the interference due to the decoded stream. For a non-dispersive overnight, the channel simulator 47 2 multiplies the modulated symbol stream of the jth transmission antenna by the vector 匕, which is between the data flowing back -67- 1269549

(63) Estimation of the channel response between the jth transmit antenna and each of the N R receive antennas. The vector L can be expressed as:

Equation (56)

A is the predicted row of the channel response matrix, which can be expressed as:

Equation (57) H =

Κι (2 A \nt h eight stone 2, NT Μ MO Μ

The ^Nk, 2^^Nft.NT matrix i can be provided by the channel/data processor 460 in the same phase. If the remodulated symbol stream for the jth transmit antenna is denoted as "χ", then the estimated interference component P of the jth reply transmitted signal can be expressed as: K)

Μ Equation (58)

KrxJ is based on the symbol stream transmitted at the jth transmit antenna, and the NR elements in the interference vector correspond to the components of the received signal at each NR receive antenna. The elements of each vector represent an estimated component of the decoded data stream due to the associated received modulated symbol stream. These components are for the interference of the remaining (not yet detected) transmitted signals in the modulated symbol stream (i.e., vector Lk), and are subtracted from the received signal vector Lk by a sum master 4 74. (ie eliminate) to provide removed has been -68 - 1269549

(64) A modified vector f+i of the components of the self-decoded data stream. This elimination can be expressed as shown in the equation (55). The modified vector Lk+1 is provided as the input vector for the next receive M processor processing stage, as shown in Figure 4E. * For a scatter channel, the vector i is replaced by an estimate of the channel transfer function vector (i(k), 0 S k S L) defined in one of the programs (23). The estimated interference vector p(η) at time η can then be expressed as:

Equation (5 9) ^Kj{k)Xj{n-k)

Jt=S〇 iA Qing

Ic^Q M .*=〇 where xj(n) is the remodulation symbol of time n. Equation (59) is effectively used to estimate the channel response for each transmit-receive antenna pair to swirl the re-modulation symbol. For simplicity, the receiver architecture shown in Figure 4E provides (received or modified) corrupted symbols to the receiver processing stages 4 5 〇, and the interference components in these streams due to previously decoded data streams have been Remove (ie remove). In a particular embodiment of Figure 4E, the stages remove interference components due to previously decoded data streams. In some other designs, the received modulated symbol stream can be provided to all stages, and each stage can be implemented to eliminate interference components from all previously decoded data streams (which can be provided by the previous stage " interference cancellation can be swept by one or Multiple stages (if the SNR of the data stream is high) Various modifications can be made to the receiver architecture shown in Figure 4E while still falling within the scope of the invention. -69- 1269549

(65) Continued Consumer Abdication in the title data transmission It is quoted 7. HJ Complete or News. Can be as below. In the case of a power-receiving channel, the SNR can also be used in another method of adding a noise power transmission channel or in another power and miscellaneous channel or in another series of receiver processing techniques. Explain in detail in the aforementioned Murray case serial number 09/854,235, and in the literature on the "V-impact: reaching a very high rate through extensive diversity of wireless channels" by pww〇lniansky, in the literature published by ISSSE-98 in Italy. The manner is incorporated herein.

Reporting complete or partial C 5U Part of the CSI may include any type of information indicating the characteristics of the communication link, or some types of information provided by CSI.

In some C SI body embodiments, part of the CSI contains SNR, which is derived as noise and interference power. The S NR is typically estimated and provides for each data transmission (e.g., each transport stream), although it is cumulatively provided to the majority of the transmission channels. The estimation of SNR is yet another. In one embodiment, the estimate of SNR will be mapped to one, for example using a lookup table. In a specific embodiment, a portion of the CSI includes a signal power and interference rate. These two components can be derived separately and provided to each of a set of transmission channels for data transmission. In one embodiment, part of the CSI includes signal power and interference power. These three components can be derived to provide a transmission channel for each transmission group for data transmission. In one embodiment, the portion c S I contains the interference power of the signal-to-noise ratio plus the observable interference. This information can be derived from 1269549 (66) for each transmission channel or a set of transmission channels for data transmission.

In yet another embodiment, a portion of cs I includes a matrix type signal component (eg, a NR X Ντ composite input for all transmit-receive antenna pairs), and a matrix type of noise plus interference component (eg, Nr χ Ντ composite) Enter). The transmitter unit can then correctly combine the signal components with the noise plus interference components and the appropriate transmit-receive antenna pair to derive the quality of each transmission channel for data transmission (eg, for each transmitted data stream received at the receiver unit) Post processing SNR). In yet another embodiment

A data transmission rate indicator of the flow. The transmission channel that will be used for data transmission is determined at the beginning (for example, based on the S ν r of the transmission channel), and the data transmission rate of the determined channel quality can then be identified (eg, a questionnaire). . The identified data transmission rate is the maximum data transmission rate 指示 indication that can transmit the required performance level on the transmission channel. The data word rate can be mapped to and represented by a Data Rate Indicator (DRI), which can be effectively encoded. For example, (5) fruit (up to) seven available data transmission rates are supported by the transmitter antennas of the transmitter unit, then a 3 _ bit

The value will be used to represent the seven Xuan D in DRI where, for example, zero can represent zero data transfer rate (ie, no use of the transmission teleport, ^, 迗天, '泉) 叩 1 to 7 can be used to represent seven different Data transfer rate. In a typical per ambulance ^ ]

褚#, ,, soil κ, her channel quality (such as SNI estimate) can be directly mapped to (4) according to, for example, the lookup table. - In yet another embodiment, the portion csi contains an indicator that will be used to communicate a particular processing scheme for each transport stream. In this specific implementation, the indicator identifies the particular code and tone that will be used to transmit the data stream.

1269549 (67) Plan to achieve ^ ^ < ten energy level * $ ° In yet another embodiment, part C s 1 contains different indicators of the quality of the particular product used for the transmission channel. Initially, the SNR or DRI of the transmitted wanted is determined by the same reference metric as the quality of some other products. After r, the quality of the transmission channel is continuously monitored and then the difference between the last reported metric and the current metric is determined. The difference can then be quantized into one or more bits, and the quantized difference is mapped to and represented by a differential indicator, which is then reported. The difference indicator can indicate the increase or decrease of the last reported amount of lunar value (or maintain the last reported metric value) in a specific level. For example, the difference indicator may indicate (1) that the apparent SNR for a particular transmission channel has increased or decreased by a particular level size' or (2) that a particular amount of data transmission rate should be adjusted, or some other change. The reference metric values may be transmitted periodically to ensure that errors within the differential metrics and/or errors in these metrics are not cumulatively received.

Complete CSI In a specific embodiment, the complete CSI contains an intrinsic wedge plus any other information that is expressed or equal to SNR. For example, the SNR correlation times: -, the information can be the data transmission rate indication of the eigenmode, the indication of the coding and modulation scheme to be used for each ancestor mode, and the dry % of each eigenmode Interference power, signal-to-interference ratio for each eigenmode, and so on. Some of the CSI information on the way up can also be provided as SNR related information.

In another embodiment, the complete C S I package, c. H ^ port a matrix! = Move η迀. This matrix is sufficient to determine the inherent mode and characteristics of the channel and can be a more efficient representation of the channel (for example, fewer bits may be needed to pass the full csi). -72- 1269549 (68) The pwmmm difference update technique can also be used for all complete c s i data types. For example, a difference update of a complete c s I feature can be sent periodically, when the channel changes some number, and so on. All or part of the CSI of other types may also be used without departing from the scope of the invention. In general, full or partial CSI includes sufficient information to be used in any form of adjustment processing at the transmitter unit so that the level of performance required to transmit the data stream can be achieved.

The derivation and reporting CSI can derive CSI based on the signals transmitted by the transmitter unit and received at the receiver unit. In a specific embodiment, the CSI is derived from a preamble included in the transmitted signal. Alternatively or additionally, the CSI can be derived from the information included in the transmitted signal.

In another embodiment, the CSI includes one or more signals transmitted on the reverse link from the receiver unit to the transmitter unit. In some systems, the association may exist between the downlink and the uplink (eg, a time division duplex (TDD) system where the uplink and downlink share the same system bandwidth in a time division multiplex manner ). In these systems, the quality of the downlink can be estimated based on the quality of the uplink (to the extent necessary), which can be estimated based on the signal transmitted by the receiver unit (e.g., the pilot signal). The pilot signal transmitted on the uplink will then represent an average value by which the transmitter unit can estimate the CSI observable at the receiver unit. In a TDD system, the transmitter unit can derive a channel coefficient matrix H (eg, based on a preamble transmitted on the uplink), a difference between the transmission and reception array copies, and a pre-73 reception noise variation at the receiver unit. - 1269549

(69) Estimate. The difference between the array copies can be addressed by a periodic calibration step that can involve feedback between the receiver unit and the transmitter unit. * Signal quality can be estimated at the receiver unit based on various techniques. Some of the techniques are described in the following patents, which are assigned to the assignee of the present invention, which is hereby incorporated by reference: • U.S. Patent No. 5,799,005, issued Aug. 25, 1998, title "System and method for determining the reception of pilot power and path loss in a CDMA communication system"; φ • US Patent No. 5,903,554, issued May 1, 1999, entitled "Used in a Spread Spectrum" Method and apparatus for measuring link quality in a communication system; • U.S. Patent Nos. 5,056,109 and 5,265,119, issued October 8, 1991 and January 23, 1993, respectively, entitled " Method and apparatus for controlling transmission power in a CDMA cellular mobile telephone system": and

• U.S. Patent No. 6,097,972, issued August 1, 2000, entitled "Method and Apparatus for Processing Power Control Signals in a CDMA Mobile Telephone System." C S I can be reported to the transmitter unit using various C S I transmission schemes. For example, C SI can be sent in whole, differentially, or a combination thereof. In a specific embodiment, the full or partial C S I is periodically reported, and the difference update is sent based on the previously transmitted CSI. As in the example of a complete CSI, the updated data may be a correction to the reported inherent mode (based on an error signal). This trait value is usually not as fast as the eigenmode changes, so these can be compared to -74 - (70) 1269549

Low rate update. In another embodiment, the CSI is sent only when there is a change (eg, when the change exceeds a certain threshold), which may decrease the effective rate of the low feedback channel, such as a portion of cs; This SNR is only sent back when r· is changed (eg differentially). For a FDM system (with or without ΜΙΜΟ), the frequency domain association can be used to allow the number of csi to be returned to be reduced. As with the partial cs; [the example of the OFDM system, if the snr system corresponding to a particular spatial subchannel for the nm frequency subchannel is similar, the SNR and the first and last frequency subchannels when this condition is true φ can be reported. Other compression and feedback channel error recovery techniques that reduce the amount of CSI data feedback may also be used and fall within the scope of the present invention. Various types of information for use in CSI and various CS] [reporting mechanisms are described in US Patent Application Serial No. 8/963,386, filed on November 3, 1997, entitled, And the device, which is assigned to the assignee of the present application, and in the "TIE/ELA/IS_856 cdma2000 high-rate packet data wireless interface specification",

Using one of the CSI (e.g., CCMI, MMSE, MMSE-LE and DFE) techniques described herein or one of the complete CSI techniques, complete or partial CSI for each transmission channel for receiving signals can be obtained. The complete or partial CSI for the transmission channel is then reported back to the transmitter unit via the reverse channel. For some C SI technologies, an adaptive process can be achieved without the need for a complete CSI. For complete CSI techniques, sufficient information (and no explicit feature values and eigenmodes) will be fed back to the transmitter unit to help calculate the eigenvalues and eigenmodes used by each frequency sub-pass. By reciprocating the CSI, it will be -75- 1269549

(71) Adaptive processing (eg, adaptive coding and modulation) can be implemented to improve the usability of the channel. Referring back to Figure 2A, on the downlink, the complete or partial CSI (eg, channel SNR) determined by the rX ΜΙΜΟ processor 260 will be provided to a data location, processor 2 80, which processes c SI The processed data is provided to one or more modulators 254. The mutator 2 5 4 further adjusts the processed data and transmits c s I back to the base station via the uplink. At base station 104, the transmitted feedback signal is received by antenna 224, demodulated by demodulator 222, and provided to an rX ΜΙΜΟ/data processor 240. The RX/data processor 240 performs processing complementary to the implementation at the data processor 280 and replies to the reported complete/partial CSI, which is then provided to the data processor 210 and the processor 220. And it is used to adjust the process. The base station 104 can adjust (i.e., adapt) its processing in accordance with the full or partial CSI from the terminal unit 106. For example, the decoding for each transmission channel can be adjusted to match the information bit rate to the transmission capability supported by channel S N R . In addition, the modulation scheme of the transmission channel can be selected according to the channel SNR. Other processing (e.g., interleaving) can also be adjusted to fall within the scope of the present invention. The adjustment of the processing of each transmission channel based on the determined channel SNR allows the system to achieve high performance (ie, high throughput or bit rate to a specific performance level). Adaptive transmission processing can be applied to a single carrier ΜΙΜ0 system or Multi-carrier basic system (for example, a system using 0FDM). The choice of adjustment and modulation of the code at the transmitter unit can be achieved in accordance with a number of techniques. [Some of this technique is in the aforementioned U.S. Patent Application.

-76- 1269549

(72) Nos. 09/776,975, 09/532,492, and 09/854,235. Part of the CSI technology (eg CCMI, MMSE, MMSE-LE and DFE technology) · With the complete CSI technology, a receiver processing technology that allows a single system to use the additional range generated by the use of multiple transmit and receive antennas Lee 1 has a major advantage. Some CSI techniques allow the same number of modulation symbols as the full CSI MEMO system to be used for transmission in each time slot. However, receiver processing techniques can also be used in conjunction with the full/partial C S I techniques described herein while still falling within the scope of the present invention. Similarly, Figures 4B through 4E represent four specific embodiments of the receiver unit that are capable of processing the transmission of ΜΙΜΟ, determining the characteristics of the transmission channel (e.g., SNR), and reporting the complete or partial Cs I back to the transmitter unit. Other techniques that are designed to occur with other receiver processing techniques are considered to be within the scope of the present invention. IV. Adaptation and Reuse A feature of the present invention provides techniques for (i) dividing and distributing available system resources (e.g., frequency spectrum) between cells of the system, and (2) allocating resources within each cell to the terminal for data transmission. The ability to dynamically and/or adaptively allocate resources to cells and the ability of cells to intelligently allocate resources to the terminal allows the system to achieve high levels of efficiency and performance. In the repair-re-use system, the "channel" that can be used by one terminal in one cell can be reused by another cell only when another cell also has the same channel reuse mode. For example, consider a 3-cell-containing cluster containing cells 1, 2, and 3. In this protocol, different channel groups are assigned to each cell located in this first reuse cluster. Each channel can be a one-time slot in the TDM system, a coding channel in the CDM system, and an FDM/OFDM system.

-77- 1269549 _ (73) I 爹 镍 明 续 续 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一The channels assigned to the group that repeatedly use any of the cells within the cluster are orthogonal to the channels assigned to other groups of other cells within the cluster. The cluster is then repeated throughout the network in some specific way. This strategy reduces or eliminates the common interference caused by the terminals in the cluster. While the revision-reuse method can be used to maximize the percentage of the terminal function to meet the minimum demand SNR, it is still generally insufficient because it utilizes a high reuse factor.

Figure 6A shows an example of a cumulative distribution function (CDF) of a terminal in a system that achieves SNR, based on a number of reuse patterns obtained by randomly distributing the analog terminals over the entire coverage. The X of the horizontal axis represents the SNR, and the vertical axis represents the S NR of a particular terminal that is less than the value shown in the horizontal axis, i.e., the probability of P (SNR < X). As shown in Figure 6A, virtually no terminal achieves an SNR below 0 dB. Figure 6A also shows that the probability of a larger SNR increases with greater reuse. Therefore, the P (SNR > X) of the 7-cell reuse pattern is greater than the P (SNR > X) of the 1-cell reuse pattern.

The SNR CDF in Figure 6A can be used to characterize the potential performance of the system. As in the example, assume that an SNR of at least 10 dB is required to meet the minimum instantaneous bit rate of 1 Mbps in 99.99% of the time. Using the reuse factor of one (i.e., N R E U S E = 1, each cell uses the same channel), the probability of not achieving the required performance (ie, the probability of interruption) is nearly 12%. Similarly, the cell re-use factors of 3, 4, and 7 correspond to interrupt susceptibility of 5.4%, 3.4%, and 1.1%, respectively. Therefore, in order to achieve a 10 dB SNR for 99% of the terminals, the re-use factor is at least 柒 (NREUSE 2 7) in this example. Figure 6B shows that the terminal in the cell of the 1-cell re-use mode reaches -78 - 1269549 _ (74) I

An example of SNR CDF. For the uplink, the SNR CDF in Figure 6B is achieved at the base station for a full-power transmission on each channel within each cell. For the downlink, the SNR CDF can be reached at the terminal when all cells are transmitting at full power. In both cases, the terminal system is unevenly distributed (i.e., randomly located) within the cell. The SNR CDF provides a percentage indicating that the intracellular terminal has an SNR greater than a particular SNR value. From Figure 6B, it can be seen that terminals with different SNR characteristics in the cell and possibly reaching different performance levels or specific performance levels may need to be transmitted at different power levels. Terminals with less path loss to the serving cell typically have a higher SNR, which means that a higher throughput will be achieved. In a typical system, a larger percentage of the terminals in the system are able to achieve an SNR equal to or greater than the set point. The set point is a specific SNR that needs to achieve the required performance level, which can be quantified as a specific average data transmission as in 1% BER or 0.01% interrupt probability or some other standard.

rate. For these terminals, the repeated use mode of one (NREUSE = 1) can be used to achieve high system efficiency. Only a portion of the terminals in the system are usually at a disadvantage. For terminals where the SNR reaches below the set point, certain other re-use schemes and/or techniques may be utilized to provide the required performance. Adaptive reuse will be provided herein to dynamically and/or adaptively allocate and allocate available system resources to cells, based on a number of factors such as observable loading conditions, system needs, and the like. The reuse plan is defined at the outset, and each cell line is assigned a portion of the overall available system resources. Allocating allows the cells to use most of the available resources at the same time, such as -79 - 1269549 (75) required or required. As the system changes, the reuse plan can be redefined to reflect the changes in the system. In this way, an adaptive reuse plan can achieve a very inefficient reuse factor (for example, close to 1) to meet other system requirements.

System resources can also be divided such that each cell can be assigned to a set of channels with different performance levels. Higher performance can be achieved for, for example, a lightly shared channel and/or a low transmission power level in conjunction with adjacent cells. Conversely, lower performance can result from, for example, low transmit power levels of the channel. Channels with different performance levels can be obtained by defining different backoff factors for the channel, as described below. On the uplink, the terminals in each cell are assigned to the channel based on the interference tolerance level of the terminal and the performance of the channel. For example, a poor terminal that requires better interference prevention can be assigned to a channel that provides more protection. Conversely, an advantage terminal with favorable propagation conditions can be assigned to channels that are more heavily shared and/or have relatively large interference levels when used.

Figure 6C is a diagram of a specific embodiment of resource partitioning and allocation for a 3-cell reuse pattern (i.e., NREUSE = 3). In this example, system resources are divided into 12 segments. The distinction can be implemented in time, frequency or coding domains or a combination thereof. Thus, the horizontal axis in Figure 6C can represent one of time or frequency, depending on whether TDM or FDM/OFDM is utilized. For example, the 12 segments may represent 12 time-division multi-time slot pairs of a TDM basic scheme or 12 frequency bands of an FDM basic scheme. Each segment may also be referred to herein as a "channel" and each channel is orthogonal to the other channels. For the 3-cell re-use mode, system resources can be divided by aggregating the available channels -80-1269549 (76), and the cells in the 3-cell cluster can be assigned a channel. group. Each channel group includes some or all of the 12 available channels, depending on the particular reuse scheme being utilized. For the specific embodiment shown in Figure 6C, each cell is assigned an equal number of channels, cell 1 is assigned channels 1 through 4, cell 2 is assigned channels 5 through 8, and cell 3 is assigned channels 9 through 12 . In some other specific embodiments, each cell can be assigned a set of channels each of which can include any number of channels, some of which can also be assigned to other cells.

1. Adaptive reuse scenarios Adaptive reuse scenarios can be designed to take advantage of certain features of communication systems to achieve high system performance. These system characteristics include the loading effect and the tolerance of the terminal to different disturbances.

The load at the cell affects the overall performance of the system (eg flux). At low loads, the available system resources can be divided into "orthogonal" channel groups, which can then be assigned to cells, one channel per cell in the cluster. Since the channels in each group are orthogonal to the channels of the other groups, the interference on these orthogonal channels is lower, and a high C/Ι value can be achieved. As the load increases, the number of orthogonal channels in each group may not be sufficient to meet the demand, and the cells may be allowed to detach from the limitations of using only the orthogonal channels. Transmission on non-orthogonal channels increases the average interference level observed in the used channel. However, by properly controlling the transmission level on the non-orthogonal channel, the amount of interference can be controlled to achieve high performance even at higher loads. As the load increases, the number of active terminals that require data transmission also increases, and a cell that may select a schedule for data transmission and a bad terminal assigned to the channel also increases. Each bad terminal is presented to the other end of the system -81 - 1269549 _ (77) I issued a boat reading page

The interference of the terminal, which can be (partially) depends on the terminal to the cell in the service, and the relative position of the other neighboring cells to the terminal. Terminals with large link limits have a large tolerance for interference. Different interference characteristics of the terminal can be used in the scheduled terminal and the designated channel for compact reuse (i.e., close to 1). In particular, when the load is increased, a terminal having a higher tolerance for interference can be assigned to a channel having a higher probability of receiving a high interference level.

Figure 7 is a flow diagram of a particular embodiment of a process 700 for an adaptive reuse scenario. The adaptability of the re-use plan and the reuse plan to system condition changes can be implemented in conjunction with the normal operation of the system.

Initially, the system is classified in accordance with one or more parameters and based on information stored in a database 730 for collection by the system. For example, as observed at each cell (for the uplink), or at each terminal (for the downlink), the interference experienced by the terminal can be determined, and an interference feature classification can be generated. The interference feature classification can be implemented on a cell basis and can participate in a statistical feature classification that produces interference levels, such as power distribution. Information for feature classification can be updated periodically to take into account new cells and terminals and to reflect changes within the system. In step 7 1 2, the reuse plan can be defined using the resulting system feature classification and other system limitations and considerations. The reuse plan contains various components, such as a specific reuse factor NREUSE and a specific reuse cell configuration based on the reuse factor NreuSE. For example, the re-use factor can be in accordance with a cell, a 3-cell, a 7-cell or a 19-cell re-use pattern or cluster. The choice of re-use factor and the design of the reusable cell configuration can be achieved according to the data generated in step 7 10 0 and any other available data. This reuse plan provides a system for the operating system.

Additional parameter parameters and/or operating conditions may also be defined in step 71. This typically involves dividing the entire available system resource into a channel, with the channel corresponding to a time unit, a frequency sub-channel, a coded channel, or some other unit. The number of channels Nch to be utilized can be determined in accordance with the reuse plan defined in step 712. The available channels will then be associated to each group and each cell will be assigned a respective channel group. The group may include overlapping channels (i.e., a particular channel may be included in more than one group). Resource allocation and allocation will be further detailed below. Other parameters can also be defined in step 7.1, such as the transmission interval, the set point of cells in the system, the backoff factor associated with the assigned channel, the limit of the backoff factor, the size of the adjustment factor, and others. The backoff factor determines the reduction in the peak transmit power level of the channel. These parameters and conditions, which are further detailed below, are a set of operational rules that cells will follow under normal operation.

The system then operates (e.g., schedules) based on the defined reuse plan and the transfer of data to the cells and/or terminals. In operation, system performance will be evaluated in accordance with the defined reuse plan in step 716. This assessment may include, for example, determining the effective path loss between each terminal and several nearby cells, as well as a measure of the associated link limits, throughput, interruption probability, and other performance. For example, the effective link limits of each of the scheduled terminals in each channel of each cell can be determined. Based on the calculated link limits, an estimate of the system's average throughput and individual performance of the terminal can be generated. Once the system performance has been evaluated, it will be defined in step 7 1 8 -83 - 1269549 (79)

The decision on the efficiency (ie performance) of the reuse plan. If system performance is not accepted, the process will revert to step 7 1 2 and the reuse plan will be defined again. System performance may not be acceptable if it does not meet a set of system requirements and/or performance levels that do not meet the requirements. The redefined usage plan may include changes to various operational parameters, and may even include selecting another reuse mode and/or reusing the cellular configuration. For example, if subjected to excessive interference, the reuse pattern can be increased (e.g., from 3-cell to 7-cell). Steps 7 1 2 to 7 1 8 can be alternately implemented until the system goal is achieved (e.g., maximizing throughput while meeting the minimum performance requirements of the terminal within coverage). Steps 7 1 2 to 7 1 8 also represent an ongoing process during system operation. If system performance is acceptable (ie, meets system requirements), then a decision will be made in step 702 regardless of whether the system has changed. If it has not changed, the processing will terminate. Otherwise, database 73 0 will be updated at step 724 to reflect changes within the system and the system will be reclassified.

The process shown in Figure 7 can be performed periodically or upon detection of a system change. For example, processing can be performed as the system grows or changes, such as when new cells and terminals are added and when existing cells and terminals are removed or modified. This program allows the system to adapt to changes (for example, in terminal distribution, topology, and terrain). 2. Power Rebate According to one feature of the invention, a channel configuration can be defined and utilized by the system such that when the load increases, the channel can be used for a greater percentage of time to achieve reliable performance. For a particular cell, it is possible that some terminals are more immune to interference from other cells or other terminals than the -84-1269549 (80) pmmMm terminal. By providing a one-pass configuration with this advantage, system throughput and performance improvements will be achieved.

For this channel configuration, each cell within the reuse cluster is assigned a respective channel group, which can then be assigned a terminal within its coverage. Each cell is further assigned a set of concession factors for a set of distribution channels. The backoff factor for each assigned channel represents the maximum percentage of total transmit power available to the channel. The backoff factor can be any value between zero (〇·〇) and one (1.0), where zero means no data is allowed to be transmitted on the channel and one means that data is transmitted at up to full transmit power. This backoff factor causes the channel to reach different performance levels.

The rebate from full transmit power can be applied to one or more selected channels, one or more selected time slots, one or more selected cells, or any combination thereof. The concession can be applied to the selected terminal in the cell either externally or selectively. In a specific embodiment, each cell applies a concession for each designated channel for data transmission, wherein the specific value for the concession is based on the operating conditions of the cell, which allows the performance of the demand to be achieved and is limited to other intracellular terminals. Interference. The channel backoff factor for assigning to each cell can be determined based on a number of factors such as the characteristics of the terminal, the loading conditions at the cell, the demand performance, and the like. The set of concession factors assigned to each cell can be unique or can be shared between different cells within the system. Typically, the channels assigned to each cell and the specified concession factor can be dynamically and/or adaptively varied depending, for example, on operating conditions (e.g., system load). -85 - 1269549 _ (81) I 瞵 _ 续 ^ ^ ^ 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在. These terminal inconsistencies can be applied according to their profiles as described below. This weighting can be adaptive and/or dynamic, such as time-date dependent.

The SNR of a particular terminal depends on various factors, including (1) path loss between the terminal and the service (or source) cells and (2) interference levels of other cells or other terminals. In a revision-terminal system, the path loss of a terminal does not change significantly and the signal level of the terminal can be accurately predicted. On the downlink, the interference level depends on the path loss from other interfering cells to the terminal, as well as the path loss from the serving cell. In the uplink, the level of interference depends on the path loss from other interfering terminals to its serving cells, as well as the path loss from these terminals to the cells they care about. Accurately predicting the level of interference typically requires that the cell or terminal is transmitting a momentary knowledge of its power level.

Many assumptions can be made to simplify the interference characteristics on the downlink and uplink. On the downlink, it can be assumed that interfering cells operate at full power. On the uplink, one of the terminals in each cell can be allowed to transmit on each channel allocated to the cell. In the worst case, the interference level of the other terminal can be based on the assumption that the terminal will transmit at full power. Decide. In contrast, the worst-case SNR of each terminal in each cell can be estimated in the worst case. The interference level of other terminals can be estimated based on the assumption that the terminal and other interfering terminals will transmit at full power. The SNR values of the terminals within each cell can be collected and used to characterize a valid SNR CDF of the cells. To derive the channel's backoff factor, the terminals in each cell can be classified according to their link limits, and the backoff factor can then be selected based on the link limit type. Using the example of the S N R distribution shown in Figure 6 B, the total number of terminals can be grouped into groups, with each group including terminals that experience similar interference levels (i.e., have SNRs within a range of values). As an example, the C D F shown in Figure 6B can be divided into Neh groups, where the Nch is assigned to the total number of channels per cell. These groups can be selected to be the same size (i.e., each group includes terminals of the same percentage), however, divisions of groups of different sizes can also be defined.

Table 3 represents the group of Neh=12 terminals and (field 2) lists the minimum S N R of the terminals in each of the 12 terminal units. Since there are 12 terminal units and each group is the same size, each group includes nearly 8.3% of the terminals in the cell. The first group includes terminals with SNR of 10 dB or less, the second group includes terminals with SNR between 10 dB and 13 dB, and the third group includes terminals with between 13 dB and 15 dB. The SNR and so on, while the last set of terminals included has an SNR greater than 34.5 dB.

Table 3 Minimum SNR range of the terminal unit _ s(n) (dB) /3(η) 1 <10 < -5 1.0000 2 10 -5 1,0000 3 13 -2 1.0000 4 15 0 1.0000 5 17 2 0.6310 6 18.5 3.5 0.4467 7 20.5 5.5 0.2818 8 22 7 0.1995 9 24 9 0.1259 10 26 11 0.0794 -87 - 1269549 (83)

_蹲 Nickel_Read Buy I :::::::::::::3⁄43⁄43⁄43⁄43⁄43⁄43⁄43⁄4^ 11 29.5 14.5 0.0355 12 > 34.5 > 19.5 0.0112

The cells can be designed to support a particular set point Ysp (or operating point) that is the minimum requirement S N R that can be manipulated at a desired data rate at an acceptable error rate. In a typical system, the set point is a function of the instantaneous data rate selected by the terminal and can therefore vary between a terminal and another terminal. As a simple example, assume that a 15 dB set point is required for all terminals in the cell. The minimum link limit s(n) of each group of terminals can be calculated as: s(n) = min{SNR(n)} - γ5ρ where η = 1,2,...,Nch Equation (60)

The minimum key limit s(n) of each group of terminals is the difference between the minimum SNR of the terminal and the set point Ysp in the group. Based on the assumption that all terminals in the system are full transmit power, the minimum link limit s(n) represents the value of the demand transfer power that deviates from the set point. A positive link limit indicates that the SNR system is greater than the required performance level defined by the set point. Therefore, the transmit power of these terminals can be reduced (ie, concessed) proportional to the link limit while still providing the required performance level. The concession factor for each cell can then be derived from messages such as path loss between the terminal and the cell and the characteristics of the interference level. If the maximum transmit power level is constantly normalized to 1.0, the normalized backoff factor for each group of terminals can be expressed as: β(η) = min(1.0, 10·0 1 s(n)) where η = 1, 2,...,Nch Equation (61) A specific terminal unit backoff factor represents the reduction in transmission power, which is -88-1269549

(84) Can be applied to the group of terminals while still maintaining the required setpoint Ysp and the required performance level. The consignment of the transmission power is feasible because these terminals can achieve a better s N R . By reducing the transmission power of the terminal by the backoff factor, the interference of the terminal to other terminals can be reduced without impeding the performance of the terminal.

Table 3 lists the minimum link limits s(n) (field 3) and the backoff factor (field 4) for each group of terminals when the set point γ5ρ is 15 dB. As shown in Table 3, channels 1 through 4 have a link limit of 0 dB or less and have a gradual link limit with channels 5 through 12. Thus, channels 1 through 4 operate at full power, while channels 5 through 12 operate at decreasing power. The backoff factor can be used for the transmission of the terminal of the relevant terminal unit. For example, since Group 5 terminals have an SNR of 17 dB or better and the minimum link limit s(n) is 2 dB, the transmit power of these terminals can be withdrawn to 63.1% of the peak transmit power.

For terminals with SNR below the set point γ5ρ, there are many options available. The data transmission rate transmitted by these terminals can be reduced to be supported by SNR. Alternatively, an interfering terminal or cell that causes a low SNR may be required (temporarily) to reduce its transmit power or stop transmitting on its affected channel until the low SNR terminal can provide a service that meets the demand. In a specific embodiment, once the factor of the cell in the reuse mode is determined, the yield factor of the other cells in the reuse mode can be staggered, for example, for a 12 channel operation, REUS Ε = 3 (ie, 3 - Cell) Reuse mode and using Nch = 4 channel offset, cell 2's backoff factor can be shifted by 4 modulus - Neh, while cell 3's backoff factor can be shifted by 8 modulus - Nch. For this re-use mode, Cell 1 applies a concession factor to link the first channel group (which includes -89-1269549 hairs to indicate the performance page (85) channel and the concession factor as shown in block 4 in Table 3), cell 2 application The backoff factor links the second channel group (which includes the channel and the backoff factor as shown in column 4 in Table 3, but moves down 4 channels and back), while cell 3 applies the backoff factor to the third channel group (which includes The channel is with the backoff factor shown in 4 position 4 in Table 3 but shifted down by 8 channels and back). A 4-channel offset is used in this example, but other offsets can be used.

Table 4 lists the concession factors for cells 1 to 3 using the concession factor shown in Table 3 and a 4-channel shift. For example, for channel 1, cell 1 applies a concession factor that binds channel 1 of group 1, cell 2 uses a concession factor that binds to channel 1 of group 1, and cell 3 uses a concession that binds to channel 5 of group 1. Table 4 Channel η βι(η) Cell 1 βι(η) Cell 2 βι(η) Cell 3 1 1.0000 0.1259 0.6310 2 1.0000 0.0794 0.4467 3 1.0000 0,0355 0.2818 4 1.0000 0.0112 0.1995 5 0.6310 1.0000 0.1259 6 0.4467 1.0000 0.0794 7 0.2818 1.0000 0.0355 8 0.1995 1.0000 0.0112 9 0.1259 0.6310 1.0000 10 0.0794 0.4467 1.0000 11 0.0355 0.2818 1.0000 12 0.0112 0.1995 1.0000

At low loads, each cell assigns a "better" distribution channel to the terminal. For the channel assignment shown in Table 4, the terminal in cell 1 is assigned the channel -90 - 1269549

(86)

From 1 to 4, the terminals in the cell 2 are assigned channels 5 to 8, and the terminals in the cell 3 are assigned channels 9 to 12. When the load in each cell is 4 terminals or less, there is no channel from the terminal of the neighboring cell to interfere with each other (since the 12 channels are orthogonal to each other), and each terminal should be able to achieve its downward use. The set point for link and uplink transmission. When the load within a cell exceeds 4 terminals, the cell can specify certain terminals to those channels that are not orthogonal to other cells. Since this load typically varies individually within each cell, it is likely that the designated non-orthogonal channels will not be occupied by any neighboring cells. The probability of this case (i.e., the probability of π no collision) is a function of the load of each cell.

A channel structure with a concession can result in an increase in the effective limit that can be observed at all terminals in the system. The backoff factors shown in Table 4 are initially derived from the SNR CDF shown in Figure 6Β, which is based on other cells delivered at full power (for the downlink), or at full power in other cells in the cell. Generated on the assumption of (for uplink) transmission. However, when the backoff factor is applied in the staggered channel reuse mode as shown in Table 4, the SNR value actually achieved by the terminal in each cell may be greater than the minimum SNR value provided by the block 2 of Table 3, due to other The terminal in a cell or other cell can be reduced by applying a concession factor. An actual system usually does not conform to the idealized system model described above. For example, non-uniformly distributed terminals, non-uniform distribution of base stations, variable geometries and morphologies, etc., all can cause observable disturbance levels in each cell. The characterization of cells and the normalization of cell properties are usually more complex than the above (ie, the SNR CDF of cells cannot be the same). Furthermore, in each cell -91 - 1269549 _ (87) I, the interference level observed in the terminal is usually different from that observed in other cells in the cell. Therefore, more calculations may be required to normalize the effective limit to within a certain threshold level of each cell across the system.

It is derived that the concession factor for each cell may therefore differ and may not be the modulus-offset type concession factor of other cells within the reuse cluster. In addition, different set points for cells and/or channels can be used to achieve a normalized performance level if desired. Setpoints can also be changed to achieve inconsistent system performance. The effect of different C/I CDFs on the concession factor and the adjustment of the concession factor to improve the performance of the system are described in US Patent Application Serial No. 09/539,157, filed March 30, 2000, entitled Methods and apparatus for system transmission are incorporated herein by reference.

Many different protocols can be used to determine the cell's concession factor. In one scenario, the procedure for determining the backoff factor is repeated several times, and the escape factor can be adjusted for each pass to match all channels to the maximum setpoint that can be reached. In a specific embodiment, the worst-case interference level is assumed when determining the initial backoff factor. In another embodiment, other values may be used in place of the worst case interference level. For example, an average, intermediate, or 95% interference distribution can be used to determine the initial backoff factor. In yet another embodiment, the interference level is adaptively estimated, and the backoff factor is periodically adjusted to reflect the estimated interference level. The concession factor used by each cell may or may not be communicated to neighboring cells. In some embodiments, a subset of the dispensed channels in a cell can have -92-1269549

(88)

There are some forms of "protection." This protection can be achieved by retaining one or more channels on a periodic architecture for terminal use within the cell. This specificity can also be defined to be used only when needed, and to the extent that it does not meet the needs of a poor terminal. Protected channels allow adjacent cells to be identified by a variety of methods. For example, a cell can communicate a list of protected channels to its neighboring cells. The neighboring cells can then reduce or avoid the transmission of data on the protected channel by the terminal within its coverage. Channel protection can be used to service poor terminals that cannot achieve the required SNR due to excessive interference from other cells or other terminals. In these cases, channel protection can be removed once these poor terminals are used.

In some embodiments, a cell may use "blocking" (ie, not using a terminal to transmit within its coverage) on some channels if the channel condition is reduced to an unacceptable level (eg, in case FER Is above a certain percentage, or the probability of interruption exceeds a certain threshold). Each cell can measure the performance of the channel and its own barrier to implementation, until it can be reasonably determined that the channel conditions are improved and reliable communication can be achieved. The channel protection and blocking can be performed dynamically and/or adaptively depending on, for example, the conditions of the cells. The adaptive re-use and power retreat for the downlink and the uplink are further described in detail in the aforementioned U.S. Patent Application Serial No. 09/539,157, the entire disclosure of which is incorporated herein by reference. The title is "Method and Apparatus for Controlling Uplink Transmissions of a Wireless Communication System," the assignee of which is incorporated herein by reference. V· Scheduling -93 - 1269549 Sending a comment on the page (89)

By scheduling and assigning terminals to assigned channels to support the simultaneous transmission of data, various scheduling schemes can be designed to maximize system throughput. A scheduler can be used to evaluate any specific combination of terminals to provide optimal system performance (eg, maximum throughput) for any system limitations and needs. By employing multiple user diversity, the scheduler can discover a combination of "compatible" terminals for simultaneously transmitting data on the assigned channels. For a system, the average system throughput can also be increased by using the “space signing” (and the feasible frequency signature) of individual terminals (ie, the channel response estimate).

The data can be transmitted by scheduling terminals according to various factors. A set of factors can be related to system constraints and requirements, such as quality of service (QoS) of demand, maximum potential factor, average data transfer rate, and so on. Some or all of these factors may need to be met on each terminal architecture within a multi-directional proximity system (ie for each terminal) and another set of factors may be related to system performance, which may be achieved by averaging system throughput The rate or some other indication of performance is quantified. These various factors will be further detailed later. The schedule can be implemented in each transmission interval, which can be defined as any time interval (e.g., a frame or a majority frame). Cells within the system may operate according to an adaptive reuse plan (stylized as described above) and in accordance with specified rules and conditions. Under normal operation, each cell receives demand from many terminals within the cell for data transmission. The cells then schedule the terminal for data transmission to meet the goals and needs. Scheduling can be performed at each cell (ie for distributed scheduling), with a central scheduler (ie for centralized scheduling) or a hybrid solution, some of which -94- v 1269549

(90) Cell scheduling is its own transmission and a central scheduler schedule is used for the transmission of a group of cells. In the following, the initially described schedule is used in a system in which the terminal operates in SISO mode. The scheduling for single-user and multi-user models and mixed modes will be detailed later. 1 · Parameters for scheduling terminals and specified channels

Various factors can be considered when the scheduling terminal is used for data transmission and designated channels to the terminal. These factors include (1) one or more channel metrics, (2) the authority to specify the terminal in use, and (3) the relevant criteria and other factors. One or more channel metrics can be used to schedule the terminal and/or specify a channel. This channel metric may include metrics based on flux, interference, interrupt probability or some other measurement. An example of a channel metric that represents "advantages" is detailed below. However, it should be understood that other channel metrics can also be programmed and fall within the scope of the present invention.

The channel metric for a known terminal can be based on various factors such as (1) path loss of the terminal, (2) all available transmit power, (3) interference characteristics, (4) backoff factor and other possibilities. In a specific embodiment, a channel metric dm(n,k) for an active terminal can be defined as follows:

Pm (η) is a concession factor for the channel η of the cell m, where 0 < β € 1. (When β m(n) = 0, it is equal to avoiding the use of channel η by cell m); Pmax(k) is used for the maximum transmission power of terminal k;

Gm(k) is the path loss between terminal k and cell m; -95 - 1269549

(91)

Im(n) is the observable interference power at channel η of cell m; and f(x) is a function describing the "advantage" of argument X, where X is proportional to SNR.

For the uplink, the correct calculation of the interference Im(η) requires information from the interfering terminals (i.e., assigned to the same channel η) to their serving cells, and the path loss to the cell m. If power control is used, the path loss to the serving cell determines the amount of power that will be transmitted by the interfering terminal. The path loss to the cell m determines the amount of power transmitted by the interfering terminal, which treats the received cell m as interference. It is usually impractical to directly calculate other-cell interferences Im(n), since information about interfering terminals is usually not available (for example, these terminals are scheduled and specified in almost the same time by other cells) and are used for The path loss characteristics of these terminals are often inaccurate (eg, may be based on averages and do not reflect degradation).

Therefore, the interference Im(n) can be estimated according to various schemes. In an interference prediction scheme, each cell maintains statistics on the received interference power for each channel. The total received power 1 for the channel η at cell m. "11" contains power Ck(n), which receives the terminal k for scheduling in channel η, and the interference power received by other interfering terminals in other cells (plus temperature and other background noise). , other - cell interference can be predicted as: /m(n) - I〇,m(n)-Ck(n) Equation (63) where / m(n) is estimated for cells in channel ηm Other-cell interference. This other-cell interference/m(n) can be estimated for each channel and at each transmission interval to form a different-cell interference distribution for each channel. This distribution is -96 - 1269549

(92) The mean, worst case or some hundred segmentation values can then be used in equation (62) as other-cell interference Im(n). Various functions f(x) can be used for channel metrics. In a specific embodiment, the channel metric dm(n, k) represents the interruption probability of the terminal k in the cell m in the channel η. In another embodiment, the channel metric dm(n, k) represents the maximum data transmission rate that can be reliably maintained at SNR = x. Other functions may also be used for channel metrics and fall within the scope of the present invention.

The channel metric dm(n, k) can be used to schedule the terminal for data transmission or to specify a channel to the terminal, or both. When scheduling a terminal and/or specifying a channel, the channel metric can be used for each active terminal calculation for each channel within the cell. Each terminal can be tied up to a maximum of Nch values, which indicates that up to Nch can be used for the intended performance of the designated channel. For a particular terminal, the channel with the best metric can be the best channel assigned to the terminal. For example, if the channel metric dm(n,k) represents the interrupt probability, the channel with the lowest interrupt probability will be assigned to the best channel of the terminal.

The channel metric dm(n,k) can be calculated to a degree of trust according to the estimation of the parameter, including the function f (X) (for example, the path loss of the terminal k to the cell m, the interference power Im(n) observed by the cell m, etc. Wait). The value of dm(n,k) can be averaged over a period of time to improve accuracy. Fluctuations in the dm(n,k) value will likely occur due to the small signal degradation of both the signal and the interference, and the change at the source of the interference causes a change in the interference power and possibly occasional obscuration (e.g., blocking the main signal path). To account for this fluctuation, a channel with a larger backoff factor can be selected to provide a certain limit, and the data transfer rate can be adapted to the change in operating conditions. -97 - 1269549

(93) The terminal can schedule data transmissions and designated channels according to its priority order, so that higher priority terminals are generally applied before lower priority terminals. Prioritization typically results in a simple terminal scheduling and channel assignment procedure, but can also be used to ensure a fair degree of fairness between terminals, as detailed below. Terminals within each cell can be prioritized according to a number of criteria, such as average throughput, delay history of the terminal, and the like. Some of this standard will be detailed later.

In a terminal prioritization scheme, the terminals are processed in order of priority based on their average throughput. In this scenario, a "score" maintained is for each active terminal that will schedule the transmission of the data. A cell can maintain this rating for its active terminal (i.e., in a distributed control scheme), or a central controller can maintain the rating for all active terminals (i.e., with a centralized control scheme). The active state of the terminal can be established at a higher level of the communication system.

In a specific embodiment, the score (1) indicating the average flux is maintained in each of the active terminals. In one implementation, the score of terminal k at frame i (K(i) is calculated as the average flux of an index, and can be expressed as: Equation (64) (/) = flash * & (Bu 1 + & (/) / rr where (j>k(i) = 0, for i < 0, rk(i) is the data transmission rate of the terminal k in frame i (unit bit/frame) , and _ α〇 and ai are time constants used for exponential averaging. Usually rk(i) is limited to a specific maximum achievable data transmission rate rma: and a specific minimum data transmission rate (such as zero). The large α ^ value (relative to α 〇) corresponds to a longer average time constant. For example, such as α 〇 and ai are both -98 - 1269549

(94) If the system is 0.5, the current data transmission rate rk(i) is given the equal weighting as the score φ “ι-1” from the previous transmission interval. The score (t>k(i) is close to the normal state of the terminal. The average transmission rate. The data transmission rate rk(i) can be an "achievable" (ie "feasible") data transmission rate for the terminal k, according to which the terminal has been achieved (ie measured) or achievable (ie estimated) SNR. The data transmission rate of terminal k can be expressed as _TF is rk (0 = ck ^ ^°S2 0 + ) Equation (65)

Where ck is a normal number reflecting a segment of theoretical capacity, which can be achieved by a coding and modulation scheme selected for terminal k. The data transfer rate rk(i) can also be the actual data transfer rate that will be specified during the current schedule, or some other quantifiable data transfer rate. The use of data transfer rates can be introduced into the channel designation process to introduce a "stroll" effect that can improve the performance of some poor terminal devices, as detailed below.

In a further implementation, the score φ of the terminal k at frame i is used to calculate a linear average flux, which is achieved within a certain time interval and can be expressed as:

1 I (10)=7 Equation (66) The average (achievable or actual) throughput of the terminal can be calculated by covering a specific number of frames (eg covering at least 10 frames as the score. Others are expected for current use) The terminal is used to score the formula of (|)k(i) and falls within the scope of the present invention. When a terminal requires data transmission, its score is initially set to zero, and -99-1269549

(95) Then updated in each frame. As long as a terminal is not scheduled for transmission in a frame, the data transmission rate for the frame is set to zero (ie rk(1) two), and its rating is updated accordingly. If the frame is incorrectly received by a terminal, the effective data transmission rate of the terminal for the frame can be set to zero. The frame error may not be immediately known (for example due to a round-trip delay in the Ack/Nak method for data transmission), but once this information is available, the score will be adjusted accordingly.

A scheduler can use this score to prioritize the terminal for scheduling and/or channel assignment. In a particular embodiment, the set of active terminal devices are prioritized such that the lowest rated terminal is assigned the highest priority' and the most south rated terminal is assigned the lowest priority. The schedule processor can also specify a non-uniform weighting factor to the terminal score when implementing prioritized processing. This non-uniform weighting factor can take into account other factors (such as explained below) when determining the priority order of the terminal.

The priority order of the terminals can also be a function of various other factors, such as payload requirements, setpoints of achievable SNR and demand, delay experience of the terminal, interruption probability, interference to neighboring cells, interference from other cells, Data transmission rate, maximum transmission power, transmission data type, data service type, etc. A larger payload can be assigned a channel with a larger backoff factor and can be assigned a higher priority order because it is generally more difficult to schedule a larger payload for data transmission. Terminals with a higher SNR can be assigned a higher priority if higher demand for average system throughput. The priority of terminals that experience longer delays can be upgraded to ensure a -100-1269549

(96) Minimum service level. Higher priority order can be assigned to time-critical data (such as retransmitted data). The above is not an example of no leakage. Other factors are also contemplated and fall within the scope of the invention. This factor can be weighted and combined to derive the priority order of the terminal. Different weighting methods can be used depending on the setting of the optimization system target. As an example, to optimize the average flux of cells, a greater weight can be given to the terminal to achieve a SNP.

A fairness criterion can be used for scheduling terminals and designated channels to ensure (or even guarantee) a minimum service level (Go S). This fairness standard is typically applied to terminals in all systems, although a particular subset of terminals (e.g., specially crafted terminals) may be selected for fairness standards applications.

For the above-mentioned terminal prioritization scheme, the allocation of resources can be performed according to the scoring ratio. In this case, the scores of all active terminals can be correlated to the maximum terminal score to form a modified score gk(i), which can be expressed as: ^(0 = ^(〇/max{^(〇 } Equation (6 7 ) Assigning resources to a specific terminal can be based on its modified score. For example, if the terminal 1 has a rating twice that of the terminal 2, the scheduler can allocate one channel (or majority). The capacity of the channel) needs to make the data rate of the two terminals equal (this channel is available). Based on the fairness considerations, the scheduler will attempt to normalize the data transmission rate that can be used for each transmission interval. Other fairness standards It can also be employed and falls within the scope of the present invention. 2. Adaptive Reuse Schedule - 101 - 1269549 Hair Issues - Continued (97) The scheduling scheme can be implemented to incorporate the channel power limits of the structure. It can be used on the uplink and downlink channels, as described above for power backoff. On the downlink key, the terminal can be assigned a channel with the maximum power limit, which matches its selected operating mode, data. Transmission rate and setting Point. On the uplink, similar to the scheduling scheme may be used when a link having a similar terminal limits, which are designated in line with the selected mode of operation, data transfer rate and the peak power channel limits.

The system can be designed to enable power control as well as flow rate control. For downlink and uplink, maximizing throughput involves using known setpoints for different modes of operation and associated data transfer rates. When allocating resources, the scheduling scheme determines the minimum transmit power needed to support a known data rate and mode of operation. On the downlink, power adjustments can be made on a per-user basis. On the uplink, this information can be transmitted to the terminal device either explicitly or implicitly (e.g., by specifying a particular channel with a known maximum power limit).

Figure 8A is a flow chart of a specific embodiment of a program for scheduling data transmission by a scheduler based on a priority order. This prioritization-based scheduling scheme can be used for downlink or uplink, and further schedules the active terminal for data transmission based on the priority of the terminal. The particular number of terminals that can be scheduled for transmission of data in each transmission interval can be limited by the number of available channels. For example, up to Nch terminals per cell can be scheduled to be transmitted on the Neh available channel. _ Initially, the parameters for the scheduled terminal are updated before step 8 1 0. These parameters can include backoff factors, interference characteristics, path loss for the terminal, and other possibilities. This parameter can be used to determine the terminal for the above -102-1269549

(98) Channel metrics.

In step 8 1 2, the terminal then schedules and ranks in order of priority. In general, only the active terminals that require data transmission are considered for inclusion in the schedule' and then these terminals are scheduled and ranked in order of priority. The prioritization of the terminal can be implemented using any of a number of terminal-rate schemes and can be based on one or more factors (e.g., average throughput, payload, etc.). The active terminals are then ranked based on the priority of the terminal, from the highest priority to the lowest priority. At step 814, the available channel is then assigned to the active terminal. The designation of a channel usually involves many steps. First, one or more channel metrics can be calculated for each terminal of each available channel based on updated parameters. Any number of channel metrics can be used, such as one shown in equation (62). The terminal then specifies the available channels based on the terminal priority, the calculated channel metrics, and other possible factors (such as request specifications). The designation of this channel can be implemented according to various channel designation schemes (some will be detailed below).

The designation of a channel can mean that a particular channel is assigned to the data transmission and the data transmission rate that will be used. The available data transfer rates can be linked to their respective coding and modulation schemes. Each of the scheduled terminals will be able to use the appropriate coding and modulation scheme based on the specified data transmission rate (e.g., previously known). Alternatively, the encoding and modulation scheme can be transmitted to the scheduled terminal. The system parameter is then updated in step 8 1 6 to reflect the channel assignment. The system parameters to be updated may include, for example, adjusting the concession factor for the intracellular channel, according to (1) for the scheduled terminal of the cell, the passage of the -103 - 1269549 (99) thief continued to purchase the designation, (2) From other cells used to adjust the concession factor and so on. The cell may also require adjacent cells to adjust for the concession factor.

In step 8 1 8, the data is then transmitted to or received from the scheduled terminal via the designated channel. From the data transmission, the various quantities can be estimated and used for a future transmission interval, such as interference observed on each channel. In general, steps 810 through 818 are performed while the cells are being operated normally. At step 820 a decision will be made, regardless of whether another transmission interval exists. If the answer is yes, then the program reverts to step 8 1 0 and the terminal is scheduled for the next transmission. Otherwise, the program is halted at step 820. Some of these steps will be detailed later. Channel Assignment The available channels can be assigned to the active terminal, depending on various scenarios and considering various factors. These channel designations include (1) priority-based channel assignment schemes, (2) requirement-based channel assignment schemes, and (3) one-channel assignment with upgrade schemes and others.

In a channel designation party based on a priority order, the implementation channel is assigned to a terminal at a time, and the terminal with the highest priority is considered to specify the channel first, and the terminal with the lowest priority is considered. Finally specify the channel. All in-cell terminal devices begin to prioritize processing based on a number of factors, such as those described above. Figure 8B is a flow diagram of a particular embodiment of a procedure 830 for a channel designation scheme based on prioritization. Initially at step 8 3 2, the channel metric is calculated for the active terminal and the available channels. Various channel metrics can be used, such as those described above. At step 8 3 4, the current terminal -104 - 1269549 (100)

The processing and ranking are then prioritized according to the factors described above. The prioritization process can also be based on the channel metrics calculated in step 832. The terminal prioritization and channel metrics are then used to implement the designation of the channel.

At step 836, the highest priority terminal is selected from the list of active terminals and an available channel is designated in step 838. In a specific embodiment, the selected terminal system is given a first priority selection channel and is assigned an available channel with the best channel metric. In another embodiment, the selected terminal is assigned a channel with the worst metric and still meets the needs of the terminal. At step 840, the selected terminal is also assigned a specific data transmission rate based on (1) the maximum rate required by the terminal, (2) the available transmission power and the backoff factor for the designated channel, and (3) The requirements of the terminal (such as the interrupt standard).

In step 8 4 2, the designated terminal is then removed from the list of active terminals. Next, in step 844, it will be determined whether the active terminal list is blank. If the list is not blank, the program returns to step 8 3 6 and a high priority order and the most unspecified terminal will be selected for the channel. Specified. Otherwise, if all active terminals are designated channels, the program terminates 0. In a specific embodiment, if there is a tie condition when the channel is specified (eg, multiple terminals have the same or similar channel metrics), then the channel is Not specified immediately. Instead, the channels that cause the tie are marked and continue to evaluate other lower priority terminals. If the next metric of the next terminal is associated with any of the marked channels, the channel can be assigned to the terminal and removed from the list of available channels. When used for a specific end -105 - (101)1269549

The highest priority of the terminal is provided. If the pass has an additional level of ten points, the transfer rate is increased by the terminal and the terminal transfer rate. If the channel is transmitted, the data transmission and transmission interval terminal is regarded as the terminal. The first time the system is available, the demand will be increased by the end of the demand. The list of marked channels is reduced to the first order. The Dow's foot leads to the link limit of the terminal to the @M, and the data transmission rate (that is, the SNIU$ is large on the designated, 疋通迢), then (1) the data transmission rate of the terminal A r, 隹 4 bamboo 傅 羊 羊 到 到 到 到 到 到 到 到 到 到 到 到 到 到 到 到 到 到 到 到 到 到 到 到 到 到 到 到 到 到 到 到 到 到 到 到 到 到 到 到 到 到 到 到 到Increased data transmission rate (such as the core supported by the effective link limit and the flux of the system. By adjusting according to its channel designation; and/or the backoff factor, the power control can be effectively exercised in the terminal. (1) The terminal can be scheduled to transmit at a reduced (", weaker" rate" or (2) the specified information is not supported. The data transfer of the terminal can be passed through the current ("blank"), in which case the channel can be made available for another use or some other action can be implemented. The weaker or blanking priority can be increased, Enhance the opportunity that the terminal considers in the next transmission area. In the channel-specific scheme, when the specified resources can be used better, the terminal requirements or payload considerations. A specific available channel, with a smaller payload (which can satisfy the lower data transfer rate) can accept the service of the channel, and the terminal with a large payload demand (which requires the transmission rate) can be Less available available clothing .

-106- 1269549

(102) A flowchart for a request-based channel designation scheme can be implemented similarly to Figure 8B for a priority-based channel assignment scheme. In one embodiment, each of the terminals selected for the channel designation is assigned an available channel with the worst metric while still meeting the needs of the terminal. In another embodiment, the priority order of the terminal can be modified to allow the terminal with greater payload to be considered for early assignment. Many other variations are also possible and fall within the scope of the invention.

In an upgrade mode channel designation, the active terminal system first specifies the channel (as explained above based on the terminal priority or requirements) and then upgrades to a preferred channel, if any. In certain embodiments of the above aspects, higher priority terminals may be assigned a worst channel first while still meeting their needs, while preferred channels are reserved for lower priority terminals if they are needed Time. These schemes can result in continuously lowering the priority of the terminal to the preferred channel, with a associated larger yield factor being close to one (i.e., larger transmit power).

If the number of active terminals is less than the number of available channels, it allows the terminal to be upgraded to the preferred channel. A terminal can be upgraded to another unspecified channel with a higher limit than the initial designated channel. The reason for upgrading the terminal is to increase reliability and/or reduce the effective transmission power required to support the transmission. Since many unspecified channels meet the needs of the terminal, reassigning the terminal to a higher channel allows the transmission power to be reduced to this limit. Various schemes can be used to upgrade the channel, some of which are detailed later. Various other update mechanisms are possible and are within the scope of the invention. -107- 1269549

(103) In a one-lane upgrade scenario, the terminals are reassigned to the preferred available channels if they meet the needs of the terminal and provide a larger link limit. If the channel is available, the implementation of the channel upgrade first upgrades the higher priority terminals in the first order, while the lower priority terminals are upgraded later. This upgrade scheme allows some or all of the active terminals to enjoy better access with higher link limits.

Figure 8 is a flow diagram of a particular embodiment of a procedure 850 for upgrading a terminal to a preferred channel in accordance with the priority order of the terminal. The active terminal is assigned to its initial channel designation before proceeding with the channel upgrade procedure, which can be achieved using the channel assignment scheme illustrated in Figure 8B. In decision 852, it will be determined whether all available channels have been assigned to the active terminal. If all channels have been assigned, no channel is available for the upgrade and the program proceeds to step 780. Otherwise, the terminal will be upgraded to an available channel if these channels are better than the originally designated channel (i.e., associated with the preferred channel metric).

At step 854, the highest priority terminal from the list of active terminals is selected for possible channel upgrades. For selected terminals, the "best" channel from the unspecified channel list is selected in step 865. The best channel can correspond to the channel with the best channel metric available for the selected terminal. The decision made in step 8 58 will be used regardless of whether the upgrade is possible for the selected terminal. If the channel metric for the best available channel is inferior to the channel originally specified by the terminal, the upgrade will not be performed and the program will proceed to step 866. Otherwise the selected terminal will be upgraded to the best available channel at step 860. It will then be removed from the list of available channels at step 862. Originally referred to as -108 - 1269549

(104) The channel assigned to the selected terminal can be placed after the list of available channels' can be used in step 8 6 4 to assign to certain other lower priority terminals. The selected terminal is then removed from the list of active terminals in step 8.6, regardless of whether the channel upgrade is in effect.

The decision made in step 868 will be blank regardless of whether the list of active terminals is blank. If the terminal list is not blank, the program returns to step 8 5 2 and the highest priority on the list will be selected for possible channel upgrades. Otherwise, if there are no channels available for upgrade or if all active terminals have been considered. The program then proceeds to step 8 70, and the backoff factors for all channels are adjusted to reduce the transmit power of the scheduled and designated terminals. The procedure is then terminated.

The upgrade procedure in Figure 8 C effectively upgrades the active terminal to an available channel that is more likely to provide improved performance. The channel of the upgrade scheme shown in Figure 8C can be modified to provide an improved channel upgrade. For example, for a particular terminal, a channel that may be detached by a lower priority terminal is more suitable for the terminal. However, the terminal is not assigned this channel because it has been removed from the list of terminals when the lower priority terminal is considered for selection. The procedure of Figure 8C is therefore iterated many times, or other tests will be implemented to address this situation. In another alternative channel upgrade scenario, the designated terminal is upgraded based on the number of available channels. For example, if there are three available channels, each is scheduled to move to the third time slot with the specified terminal. This upgrade scheme allows most (if not all) terminals to share better channels. In a other channel designation, the channel metric associated with the channel -109 - 1269549

(105)

The difference can be considered in the channel designation. In other examples, it is best not to assign the channel with the best channel metric to the highest priority terminal. For example, many channels can associate nearly similar metrics to a particular terminal, or many can provide the required SNR. In these examples, the terminal can be assigned to one of several channels while still operating correctly. If a lower priority terminal has its best channel selected the same as the higher priority terminal, and if there is a large difference between the best and lower quality channels of the lower priority terminal, It is most desirable to specify a sub-optimal channel to a higher priority terminal and specify the best channel to the lower priority terminal. In yet another channel designation, the highest priority terminal is labeled to provide an available channel for demand performance (similar to the channel that is labeled as described above). The next lower priority terminal identifies its acceptable channel. The channel designation is then implemented such that the lower priority terminal is first assigned the channel, but the channel required by the higher priority terminal is reserved.

In yet another channel designation, by considering a larger number of transpositions between groups of active in-cell terminals within the cell, the channel will be more desirably assigned to the active terminal within the cell. In this case, the decision to specify for a particular terminal channel is not based solely on the measurement and prioritization of the terminal. In one implementation, the priority order of the terminals can be converted to a weighting that is used to rank the metrics when calculating channel assignments within the cell. Other factors can also be considered when scheduling terminals for data transmission and designated channels. First, a specific terminal can be assigned to multiple channels. If this channel is available and if one channel system does not meet the requirements of the terminal -110-1269549, the continuation page (106)

begging. Second, a particular terminal can be assigned to different channels of different transmissions to provide a "strolling" effect that provides interference averaging in some cases and can improve the performance of poor terminals. Third, the probability of other terminals transmitting on a particular channel can also be considered. If many channels have near-equal channel metrics without considering occupancy probabilities, then the channel with the lowest probability of being used in other cells can be specified. Fourth, excessive interruption probability can be considered when channel assignment is specified. If the expected interruption probability of a terminal to a particular channel exceeds the standard, then it is likely that the overall transmission on that channel will degrade and will need to be retransmitted, and it is best not to specify the channel at all or specify that channel to Another terminal that is more suitable for use.

The available channels can also be assigned to terminals with approximately no conditions or restricted use. Such conditions may include, for example, (1) data transfer rate limitations, (2) maximum transfer power, (3) set point limits, and the like. The maximum transmit power limit can be placed on certain designated channels. If the cells in the system have power data for channels in other cells, the level of interference can be calculated with a higher degree of certainty and can be better planned and scheduled. A specific set point can be applied to a specific channel, for example under high load conditions. A (e.g., low priority) terminal can be assigned a channel that does not meet the minimum interrupt probability of demand (i.e., the designated channel has an expected SNR that is lower than needed). In this case, the terminal will need to use the designated channel to operate at a lower set point that meets the required performance criteria. The set point used can be static or can be adjusted with system load. At the same time, this set point can be used on a per channel basis. -111 - (107) 1269549 β 3. Used under the system 鍅

The present invention-features provide techniques to minimize the downlink capacity of a system (e.g., a multi-directional proximity cellular system). The data may be transmitted by the base station to - or a plurality of terminals using one or more different modes of operation as described above. The available downlink resources in the single user mode can be assigned to a single terminal. In a multiple user mode (also referred to as a Ν-SΙΜΟ mode), the downlink resources can be allocated to a number of different SIMO terminals, where each terminal demodulates into a single data stream. In the hybrid mode, the downlink resources can be assigned to the combination of SIMO and the terminal, where the two types of terminals are simultaneously supported by the same channel. Multiple independent data streams can be transmitted from the base station to one or more scheduled terminals via multiple transmission days. If the propagation environment has sufficient scatter, ΜΙΜΟ receiver processing techniques can be used in the terminal to effectively develop the spatial extent of the ΜΙΜΟ channel to increase transmission capacity. Depending on the terminal, the same receiver processing technique can be used to process signals that are expected to be different for the terminal (e.g., a single terminal) or only one of the Ντ signals (i.e., siM〇 terminals). As shown in Figure 1, the terminals can be randomly distributed within the cells or can be co-located. For a wireless overnight system, its link characteristics typically change over time due to many factors such as fading and multiple paths. At a particular instant, the channel response between the NT transmit antenna array between a base station and the Nr receive antenna of a single terminal channel can be characterized as a channel response matrix, the components of which form independent Gaussian random variables, such as : -112 - 1269549 (108)

ϋ — [Ml Il2 & NT] = Κλ Κ, 2 八 At.NT 厶2,2 Λ 6 2, ΝΤ Μ Μ Μ 八^nr.n Equation (68) ^ *v where hu is coupled to the base station The jth transmitting antenna is connected to the ith receiving antenna of the terminal (i.e., (i, j) transmitting-receiving antenna pair). For simplicity, Equation (68) describes the characteristics of a channel (ie, a plurality of values for the entire system bandwidth) based on a flat decay channel pattern. In an actual operating environment, the channel can be frequency selective and a more detailed channel feature can be used (e.g., each element of the matrix iL can include a set of values for different frequency subchannels or time delays). The active terminal in the system (i.e., who is required to transmit data in a predetermined transmission interval) periodically predicts the channel response of each of the transmit-receive antenna pairs and reports the channel response CSI indication to the base station. The cumulative CSI received by the collection of active terminals can be used to (1) select the best one or more terminal units for data transmission, (2) designate available transmit antennas to selected terminals, and (3) Choose the appropriate coding and modulation scheme for each transmit antenna. With available C SI, various scheduling schemes can be designed to maximize downlink performance by providing the best system performance (eg, highest) by any combination of terminal and antenna assignments for any system limitations and needs to be evaluated. Flux). By using the "space signing" (and feasible frequency) of individual terminals (ie, the frequency of the channel), it is also possible to increase the average system throughput 0 for simplification, for various downlinks of a system without OFDM. The schedule will be detailed later, and one of the independent data streams can be transmitted from the base station.

-113 - 1269549

(109) Send antenna transmission. In this case, the (up to) Ντ independent data stream can be simultaneously transmitted by the base station from the Ντ transmit antenna to one or more of the terminals (ie Ντ X NR ΜΙΜΟ) each equipped with the NR receive antenna, where NR 2 Ντ . For the sake of simplicity, in most of the description below it is assumed that the number of receiving antennas is equal to the number of transmitting antennas (i.e., N r = Ν τ). This is not a requirement, as all analyses are applied to NR 2 ΝΤ.

The data transmission schedule on the downlink of the system consists of two parts: (1) selecting one or more group terminals for evaluation, and (2) designating the available transmission antennas to the selected terminal. All or a subset of active terminals can be considered for scheduling, and these terminals can be combined to form one or more groups to be evaluated (i.e., hypothetical). For each hypothesis, the available transmit antennas can be assigned to the terminal in the hypothesis according to any of the many antenna assignment schemes. In the best assumption, the terminal can then schedule the transmission of data in a predetermined transmission interval. The flexibility to select the best end unit for data transmission and the designated transmit antenna to the selected terminal allows the scheduler to optimize the performance of the multi-user diversity environment. To determine the "optimization" to a group of terminals, SNR or some other sufficient statistic will be provided for each terminal and each spatial subchannel. If the statistical value is SNR, then for each group to be evaluated for a terminal that transmits data in a predetermined transmission interval, a post-processing SNR hypothesis matrix for the terminal unit can be expressed as: Γ /丨", t,2 Λ ,1·ΝΤ Ύιλ Yi.1 八^2,ΝΤ Μ Μ Μ 'Νγ,1 ^Ντ>2 Λ ^ΝΤ,ΝΤ- Equation (69) -114- 1269549

(110) wherein the SNR (hypothetically) is processed after a data stream is transmitted from the jth transmitting antenna to the ith terminal.

In the multiple user mode, for a different terminal of τ, the matrix 对应 corresponds to the Ντ vector of SNR. In this mode, it is assumed that the columns in the matrix 代表 represent the SNR of each transport stream for a terminal. In hybrid mode, for a particular terminal that is designed to receive more than two streams, the SNR vector of the terminal can be copied such that the vector appears in the number of streams as will be transmitted by the terminal. In each column (ie one column for each data stream). Alternatively, it is assumed that a column within the matrix 可 can be used for each SIMO or ΜΙΜΟ terminal, and the scheduler can be designed to note and evaluate these different types of terminals.

For the terminals to be evaluated in each group, Ντ (assumed) transmission data stream is received by the NR receiving antenna of the terminal, and the > receiving signal can be processed by spatial or spatial time to be separated by Ντ as described above. Data stream. The SNR of the data stream (i.e., after spatial/spatial time processing) can be estimated and included for processing the SNR. For each terminal, a set of τ post-processing SNRs can be provided to the Ντ data stream that can be received by the terminal. If the continuous cancellation receiver processing technique is used for a terminal to process the received signal, then processing the SNR for each transmitted data stream to reach the terminal depends on the transmitted data stream being detected (ie, demodulated and decoded). The order is to reply to the transmission of the data as described above. In this case, a number of sets of S N R can be provided for each terminal in a variety of possible detection sequences. Multiple hypothesis matrices can then be formed and evaluated to determine which particular combination of terminals and detection sequences provide optimal system performance. -115 - 1269549 _ (111) In all cases, each hypothesis matrix Γ includes post-processing s N R for a particular terminal (ie, hypothesis) to be evaluated by one or more groups. These post-processing SNRs represent the SNR that can be achieved by the terminal and will be used to evaluate this assumption.

Figure 9 is a flow diagram of a particular embodiment of a program for transmitting data over a downlink system in a system. Initially in step 9 1 2, the metrics used to select the best end unit for data transmission are initialized first. Various performance metrics can be used to evaluate the terminal unit and its components will be further detailed below. For example, a performance metric that maximizes system throughput can be used.

In step 9 1 4, a set of (new) active terminals of one or more numbers are then selected by all active terminals considered for scheduling, and the set of terminals form an assumption to be evaluated. Various techniques can be used to limit the number of active terminals that are considered for scheduling, which will reduce the number of hypotheses to be evaluated, as detailed below. For each terminal in the hypothesis, the SNR vector (eg, li = [YU, Yi, 2, is captured in step 916. In the hypothesis, the SNR vector for all terminals is formed in equation (69)). The hypothesis matrix Γ. For the Γτ transmit antenna and the Ντ terminal machine hypothesis matrix Γ, there may be a combination of τ τ multiplied by the possible transmit antenna to the terminal (ie Ν τ ! sub hypothesis). Therefore, in step 9 1 8. A specific (new) antenna/terminal designation combination is selected for evaluation. The antenna/terminal designation of this particular combination forms a sub-hypothesis to be evaluated. This sub-hypothesis can then be evaluated at step 902. The metrics (such as system flux) corresponding to this sub-hypothesis will be determined (for example, based on sub-hypothesis -116 - 1269549)

(112) SNR). This performance metric will then be used in step 922 to update the performance metric corresponding to the current best sub-hypothesis. Specifically, if the performance metric used for this sub-hypothesis is better than the current best-of-life hypothesis, then this sub-hypothesis will be the most recent best sub-hypothesis, and this performance metric and other terminal metrics corresponding to this sub-hypothesis Will be stored. This performance and terminal metrics are detailed later.

The decision made in step 9.2 will be evaluated regardless of whether all sub-hypotheses used in the current sub-hypothesis have been evaluated. If not all sub-hypotheses are evaluated, the procedure returns to step 9 1 8 and a specified combination of different and unevaluated antennas/terminals will be selected for evaluation. Steps 9 1 8 to 924 will be repeated for each sub-hypothesis to be evaluated. If all of the sub-hypotheses used for a particular hypothesis are evaluated in step 924, the decision made in step 926 will be considered regardless of all hypotheses. If not all assumptions have been considered, the procedure returns to step 9 1 4 and a different and unevaluated terminal unit will be selected for evaluation. Steps 9 14 to 926 will be repeated for each hypothesis to be considered.

If all the assumptions are taken into account in step 926, then the specific end-units that are scheduled to use the transmission interval for data transmission and their designated transmitting antennas will be known. The processing SNR and its antenna designation corresponding to this set of terminals can be used to select the appropriate encoding and modulation scheme for the data stream to be transmitted to the terminal. The schedule, antenna designation, coding and modulation scheme, other information, or any combination of the foregoing may thus be transmitted to the scheduled terminal (e.g., via a control channel) at step 928. Alternatively, the terminal can implement "no destination" detection and expects to detect all transmitted data streams to determine which (if any) data stream will be used. -117- 1269549

(113) If the scheduling plan needs to maintain measurements of other systems and terminals (eg, average data transmission rates across the past K transmission intervals, potential factors for data transmission, etc.), then these metrics will be in step 93. 0. This terminal metric can be used to evaluate the performance of individual terminals and will be detailed later. This schedule is usually implemented in each transmission interval.

For a given hypothesis matrix, the scheduler evaluates the various combinations of transmit antennas and terminal pairs (ie, sub-hypotheses) to determine the best assignment for this hypothesis. Various designation schemes can be used to specify the transmit antenna to the terminal to achieve various system goals, such as fairness, performance, and the like. In an antenna designation, all possible sub-hypotheses are evaluated against a specific performance metric, and sub-hypotheses with the best performance metrics are selected. For each hypothesis matrix Γ, there will be a Ντ factor factorial (ie Ντ!) possible sub-hypothesis that can be evaluated. Each sub-hypothesis corresponds to a specific designation of each transmit antenna to a particular terminal. Each sub-hypothesis is thus represented by a post-processing SNR vector, which is expressed as: 丄 sub-hyp :a%l>Ybt27...r,NT]

Where YU is the jth transmit antenna to the ith terminal, the SNR is processed, and the subscripts {a, b, ... and r} indicate the specific terminal used for the transmit antenna/terminal pair in the hypothesis. Each sub-hypothesis will be further combined with a performance metric Rsub_hyp, which can be a function of various factors. For example, a performance metric based on post-processing S N R can be expressed as: ^•sub-hyp where f(·) is a positive real function of the index value (S) in parentheses. -118- 1269549

(114) Various functions can be used to formulate this performance metric. In a specific embodiment, a function of the achievable flux of all Ντ transmit antennas for the sub-hypothesis can be used, which can be expressed as:

Db) Equation (7〇) Μ where is the flux associated with the jth transmit antenna in the hypothesis, which can be expressed as: rj^Cj^log^l + Yj) Equation (71) where Cj is a normal number, A segment of the theoretical capacity achieved by the coding and modulation scheme selected for transmission of the data stream at the jth transmit antenna may be reflected, and γ" is used to process the SNR after the jth data stream.

The first antenna designation scheme as shown in Figure 9A, as described above, represents a specific scheme for evaluating all possible transmit antenna to terminal combinations. The total number of possible sub-hypotheses that each hypothesis will be evaluated by the scheduler is Ντ !. Since there will be a large number of assumptions that need to be evaluated, it will be a big consideration. The first scheduling scheme implements an exhaustive search to determine the sub-hypothesis that provides the "best" system performance, quantified by the performance metrics used to select the best sub-hypothesis. Many techniques can be used to reduce the complexity of the processing of a given transmit antenna. One of these techniques will be described in detail later, and other techniques may be implemented and fall within the scope of the present invention. These techniques also provide high system performance while reducing the amount of processing required to specify the transmit antenna to the terminal. In a second antenna designation, a maximum-to-maximum ("max-max") criterion is used to specify the transmit antenna to the terminal in the hypothesis to be evaluated. Using this maximum to maximum standard, each transmit antenna is assigned a specific terminal that can reach the optimum S N R for that transmit antenna. The antenna finger -119 -

1269549 (115) The system is implemented in a transmitting antenna between the same j. Figure 9B is a flow diagram of a particular embodiment of the sequence 940 of transmitting the antenna to the terminal using the maximum to maximum standard. The processing shown in Figure 2 is implemented in a specific hypothesis, which corresponds to the 〆 疋 group - or a plurality of terminals ‘ 机. Initially, in step 942, the maximum post-processing SNR is first determined after assuming that the maximum SNR corresponds to the 'specific transmit antenna/terminal pair, and the transmit antenna is at step 944. machine. The transmitting antenna and the terminal are then removed from the matrix r and the matrix is reduced to the scale of (Ντ -1) φ X (Ντ -1) by removing the row corresponding to the transmitting antenna and correspondingly at step 946. The decision made in step 948 of the terminal just designated will be regardless of whether all of the transmit antennas in the hypothesis have been designated. If all transmit antenna addresses are specified, the antenna designation will be provided in step 950 and the program will terminate. Otherwise, the procedure returns to step 924 and the other transmit antenna will be specified in a similar manner. Once the antenna of the known hypothesis matrix is completed, the performance metric corresponding to this hypothesis (eg, system flux) will be determined (eg, according to the SNR corresponding to the antenna designation), as in equations (70) and (71). Shown. This performance metric will be updated for each hypothesis. When all assumptions have been evaluated, the best combination of terminal and antenna designation will be selected to be used in the transmission interval. - Table 5 shows an example of post-processing SNR <matrix 推 derived from a 4x4 ΜΙΜΟ system terminal, where the base station includes 4 transmit antennas and each terminal includes 4 receive antennas. For the antenna finger-120 - 1269549 according to the maximum-to-maximum standard, the optimal SN R (16 dB) is obtained by transmitting the antenna 3 and assigning it to the terminal 1. The shaded columns in columns 3 and 4 of the table are shown. The transmitting antenna 3 and the terminal 1 are then removed from the matrix. The best SNR (14 dB) in the reduced 3x3 matrix is achieved by the two transmit antennas 1 and 4, which are assigned to terminals 3 and 2, respectively. The remaining transmitting antenna 2 is then assigned to the terminal 4. table 5

SNR (decibels) transmitting antenna terminal 1 2 3 4 1 7 9 iron li l liliiilll 1 boring 2 8 10 12 丨 爨 爨 round 3 1 off 丨曜 兹 兹 验 验 兹 验 验 验 验 验 验 _ _ _ _ _ Test 7 6 9 4 12 Sfi! SfifiQSfifi 7 5 _ίΙ自圆P

The scheduling scheme depicted in Figures 9A and 9B represents a particular scheme that evaluates various assumptions corresponding to various combinations of possible combinations of active terminals that are scheduled to transmit data over a transmission interval. The total number of assumptions to be evaluated by the scheduler can be quite large, even if the number of active terminals is small. In fact, the assumed total number of Nhyp can be expressed as:

N 卜)_ Nu! Ding Guang (比 - Ντ)!Ντ! Equation (72) where Nu is the number of active terminals that will be considered for scheduling. For example, if Nu = 8 and NT = 4, then Nhyp = 70. A non-missing search can be used to determine the particular hypothesis (and the particular antenna designation) that provides the optimal system performance, quantified by selecting the best hypothesis and the antenna-specified performance metric. -121 - 1269549 (117) Other quantities of processing with the performance of the South System of the scope of the invention are in the order of another one or more potential requirements), which are scheduled to be carried out * Terminal demanding (eg to Go) Update. For each schedule, it can be based on: In the embodiment, only another machine will be considered, and Figure 9C will use a specific implementation machine to be considered in the checklist. The '&amp;&gt; '^ scheduling scheme can also be implemented and is included in the book. This scheduling method, &lt; one is detailed later. These schemes also provide 'can reduce the need for I', but to be used for scheduling terminals In the data transmission, his scheduling scheme φ T 'the current terminal system is superior to the data transmission wheel according to the terminal machine. The long-term μ code goods first, the priority order of each terminal can be as measured above (such as flat and private quantity), system limitation It is derived from the demand (such as other factors or Gan, · _, , ' and is combined. All the demand for the current terminal list of 4 in the data transmission will be maintained. Winter - data transmission, He Tianyi" List, It is added to the list and its metrics will be initialized. The metrics of each terminal will be used for each transmission. Once the terminal is not raining, the data will be transmitted again, which will be in the list box π 早 _ 彳 彳 胄 or The collection may be considered for considering a specific number of terminals according to various factors. A specific highest priority terminal is selected for data transmission. In the specific embodiment, the highest priority order in the list is used for the terminal. Scheduling, where Νχ &gt; Ντ. A flowchart of an example of a prioritization-based scheduling scheme program 96, wherein a set of Ν τ highest priority terminals are scheduled. In step 962, the scheduler is in each The priority order of all active terminals in the transmission interval and the selection of the group Ντ terminal. The remaining terminals in the list will not be considered for the channel estimation of each selected terminal. In step 964 -122- 1269549

(118) Obtained. For example, processing S N R for a selected terminal can be expanded and used to form a hypothesis matrix Γ. The Ντ transmit antenna is then assigned to the selected terminal in step 966 based on channel estimation and using any of the plurality of antenna designations. For example, the antenna designation can be based on an exhaustive search or maximum to maximum standard as described above. In another antenna designation, after the terminal metric is updated,

The transmit antenna will be assigned to the terminal to keep the terminal priorities as close as possible. The data rate of the terminal and the encoding and modulation scheme are then determined in step 968 based on the antenna designation. The schedule and data transfer rate can be reported to the terminal of the #程. The schedule in the list (the metrics with the unplanned terminal are updated to reflect the scheduled data transmission (without transmission), and then the system metrics are also updated in step 970. ΜΙΜΟ System downlink scheduling The method of assigning downlink resources in a system is described in more detail in U.S. Patent Application Serial No. Re-input Multiple Turn-Out (ΜΙΜΟ) Communication 〇 9/859,345, filed May 2, 2011. And the device "to the assignee of the present invention and incorporated herein by reference.

4. Uplinks The present invention - features provide techniques for increasing the uplink key capacity of the MIM system. A scheduling scheme is provided for scheduling transmissions from the uplink data of the SIM(R) terminal, which utilizes a single antenna and/or a multi-antenna terminal. Two types of terminals can be supported on the same channel at the same time. The circle receiver processing technique can be used to process signals transmitted by any combination of SIMmm fine terminals -123 - 1269549 (119). From the point of view of the base station, there is no significant difference between the processing of N different signals by a single MIM0 terminal and the processing of a single signal by each different s I Μ 0 terminal. For simplicity, it is assumed that each terminal in the cell is equipped with a single antenna. At a particular instant, the channel response between the antenna of each terminal and the NR receiving antenna array of the base station is characterized as a vector kj whose components constitute independent Gaussian random variables such as the following:

H- M (73) where hi, is the channel response estimate from the jth terminal to the i-th receiving antenna of the base station.

Also for simplification, it is assumed that the average received power from each terminal is constantly normalized after signal processing at the base station to achieve a common set point ysp. The common set point can be achieved by adjusting a closed loop power control mechanism for each transmit terminal (e.g., based on power control commands from the base station). Alternatively, a unique set point can be used for each terminal and the techniques described herein can be generalized to cover this mode of operation. It is also assumed that the transmissions from different terminals are synchronized so that the transmission arrives at the base station within a specified time window. The base station regularly estimates the channel response of the current terminal. Based on this available channel estimate, various scheduling schemes can be devised to maximize the uplink throughput, allowing scheduling to be simultaneously transmitted by scheduling and assigning available terminals to the available transmission channels. The scheduler evaluates which particular combination of terminals can provide relative to any -124-1269549

(120) Optimal system performance (eg, maximum throughput) for system constraints and requirements. By signing the space (and the viable frequency) of the individual terminals, the average system throughput will increase relative to that achieved with a single terminal. Furthermore, by employing multiple user diversity, the scheduler can find a combination of "compatible" terminals and can be allowed to transmit on the same channel at the same time, compared to a single user schedule or for multiple uses. Random scheduling, which can effectively enhance system capacity

The uplink scheduling scheme is designed to select the best combination of terminals for simultaneous transmission over the available transmission channels to maximize system performance while still meeting system constraints and requirements. If the Ντ terminal is selected for transmission and each terminal utilizes an antenna, the channel response matrix corresponding to the selected terminal unit {ϋ = {ua,ub,...,U%} } can be expressed as: = [fei ...1ιντ ]= Κι 八V Kr Λ \ντ Μ Μ Μ ^Nr-2 Λ Equation (74)

In a specific embodiment, the continuous cancellation receiver processing technique can be used in a base station to receive and process transmissions from the terminal. When this receiver processing technique is used to process the received signal, the SNR system associated with each transmitting terminal is a function of the particular order of processing within the base station. The uplink scheduling scheme takes this into account when selecting the terminal unit for data transmission. Figure 10 is a flow diagram of a particular embodiment of a process 1000 for scheduling a terminal to transmit on the uplink. Initially, at step 1012, the metrics used to select the best end unit for data transfer are initialized first. As above -125 - 1269549 (121)

A variety of performance metrics can be used to evaluate the choice of terminal. In step 1014, a set of one or more (new) active terminals are selected by all active terminals to transmit the data in the transmissions scheduled to be made. Various techniques can be used to limit the number of active terminals that are considered for scheduling, as described above. The selected specific terminal group (eg u = {ua, ub, .., ιι~}) forms an assumption to be evaluated. For each selected terminal h in the group, the channel estimate vector ^ is retrieved in step 1016.

When using continuous cancellation receiver processing techniques on the base station, the order in which the terminals are processed directly impacts their performance. Thus in step 1018, a particular (new) order is selected to process the terminals within the group. This particular sequence forms a sub-hypothesis that will be evaluated.

This sub-hypothesis can then be evaluated at step 1020 and the terminal metrics for that sub-hypothesis will be provided. The terminal metric can be used to process SNR after signal hypothesis transmission from terminals in the group. Step 1020 can be accomplished in accordance with the above-described continuous elimination receiver processing technique. At step 1022, a performance metric (e.g., system throughput) corresponding to this sub-hypothesis is then determined (e.g., based on the terminal to process the SNR). This performance metric is then also used in step 1022 to update the performance metric corresponding to the current best sub-hypothesis. Specifically, if the performance metric used for this sub-hypothesis is better than the current best sub-hypothesis, then this sub-hypothesis will be the latest best sub-hypothesis, and the performance and terminal metrics corresponding to this sub-hypothesis will be Store. The decision made at step 1024 will be regardless of whether all sub-hypotheses used in the current hypothesis have been evaluated. If not all sub-hypotheses are evaluated, the process returns to step 1018, and a different order within the set of terminals that has not been evaluated -126 - (122) 1269549 will be selected for evaluation. Steps 1〇18 to 1〇24 will be repeated for each sub-hypothesis to be evaluated. If all sub-hypotheses about a particular hypothesis are evaluated in step 1〇24, the decision made in step 1026 will be considered regardless of whether all hypotheses are considered. If not all assumptions have been considered, then the procedure returns to step 1014 and a different and unevaluated terminal unit will be selected for evaluation. Steps 1014 through 1026 will be repeated for each hypothesis to be considered. If all the assumptions about the active terminal are taken into account in step 1026, then the result of the best sub-hypothesis will be mis-stored, and within the best sub-hypothesis, the transmission rate of the field machine will be determined (for example, Its SNR), and the subsequent scheduling and data transmission ^ ^ ^ A Feng W Feng will be in the step 1028 before the scheduled transmission of the message to,,,,,,,,,,,,,,,,,,,, The metrics (eg, the average data transfer rate across the transmissions, the potential cause of the data transfer, t, , and temple) are then updated at step 1030. The terminal musk I m . is used to evaluate the performance of individual terminals. In step 1020, you r Μ,, and again? Also &lt; The evaluation will be based on the continuous cancellation receiver processing technique described in Figure 5, if it is used in a base station. The specific order in which the receiver processes the dead, , , , ^ ' processing signals will affect the outcome. Therefore, the 闱山a receiver processing technique, for each hypothesis of the 终τ final field machine to be evaluated, such as IM, will have a possible order of Ντ factor multiplication (eg Ντ! = 24 if Ντ = 4 } Corresponding to one: the sub-hypothesis of the iNT factor factorial is used for this hypothesis. Each sub-continuation eliminates the receiver:: <final rider unit 11 = {Ua, Ub, ..., V ' and the machine (ie terminal u--then will The terminal a is processed in the specified order, followed by the terminal ub, etc.) - 127 - 1269549 嗔 嗔 戴 戴 (123) For each sub-hypothesis, the continuous cancellation receiver processing provides a set of SNRs for the The terminal then processes the signal, which can be expressed as: where γ" is the SNR of the receiver after processing the hypothesis about the jth terminal. Each sub-hypothesis will be further combined with a performance metric Rhyp^rdu, which can be various factors A function. For example, a performance metric based on the SNR of the terminal can be expressed as:

Khyp,order a f^lkyp.ordj where f(·) is a positive real number function of the arguments in parentheses. Various functions can be used to formulate this performance metric. In a specific embodiment, a function of the achievable flux of all ττ transmitting antennas for the sub-hypothesis can be used, which can be shown as shown in equations (70) and (71).

For each sub-hypothesis to be evaluated, the SNR provided by the performance elimination receiver process can be used to derive performance metrics for the sub-hypothesis, as shown in equations (70) and (7 1). The performance metrics calculated for each sub-hypothesis will be compared to the current best sub-hypothesis. If the performance metric for a current sub-hypothesis is better, then the sub-hypothesis and associated performance metrics and SNR will be stored as a measure of the most recent best hypothesis. Once all sub-hypotheses have been evaluated, the best sub-hypothesis will be selected and the terminals within the sub-hypothesis will be scheduled for transmission in a predetermined procedure. The best sub-hypothesis will be associated with a specific end unit. If the receiver processor is continuously eliminated for use at the base station, the best sub-hypothesis will be further associated with a particular receiver processing sequence at the base station. In any case, the sub-prediction will be further related to the SNR that the terminal can achieve, which can be determined in accordance with the processing order selected in the selection -128 - 1269549. The data transfer rate for the terminal can then be calculated from the s N R · achieved by it, as shown in equation (71). A portion of the cSI (which may include data, transmission rate or the s NR) may be reported to the scheduled terminal, which in turn uses the portion of the CSI to adjust (ie, rewrite) its data processing to achieve the desired Performance level.

The uplink scheduling scheme depicted in Figure 10A represents a particular scheme that evaluates all possible sequences of possible active end units that are required to transmit data in a transmission that is scheduled to proceed. The total number of possible assumptions that will be estimated by the scheduler can be quite large, even if the number of active terminals is small. In fact, the total number of sub-hypotheses can be expressed as:

NTJ—(Νυ-Ντ)! Equation (75)

The number of active terminals that will be considered for scheduling will be considered. For example, if Nu = 8 and ΝΤ = 4, then Nsub-hyp = 1680. An exhaustive search can be used to determine this sub-hypothesis that provides optimized system performance, quantified by performance metrics used to select the best sub-hypothesis. Various other uplink scheduling schemes with reduced complexity in processing scheduled terminals are also available. In this uplink scheduling scheme, the terminals included in each hypothesis will be processed in a specific order determined according to the special definition rules. In one embodiment, for each iteration, the receiver processing technique continually cancels the transmitted signal with the best SNR after equalization. In this case, the order will be determined based on the post-processing SNR of the hypothetical terminal -129 - 1269549 (125) g. In another embodiment, the terminals in each of the hypotheses are processed in a particular order. The processing sequence may be based on the priority order of the terminal in the hypothesis (for example, the terminal of the lowest priority order is processed first, the terminal of the next low priority order is processed, etc.) and the terminal of the highest priority is finally processed), the user is paid. Load, potential demand, priority of emergency services, etc.

In another uplink scheduling scheme, the scheduling of the terminals is based on the priority order of the terminals. For each frame, a specific number of terminals on the list can be considered for scheduling. In a specific embodiment, only the terminal with the highest priority of the Ντ is selected for transmission on the Ντ available transmission channel. In another embodiment, the Nx highest priority terminal on the list will be considered for scheduling 'where Νχ> Νχ &gt; Ντ.

Figure 1 OB is a flow diagram of a particular embodiment of a prioritization-based scheduling scheme program 1040 in which NT highest priority terminals are scheduled to be transmitted on the uplink. For each transmission interval, the scheduler checks the priority order of all active terminals on the list at step 1042 and selects the terminal with the highest priority of Ντ. In this particular embodiment, the remaining (Nu - Ντ) terminals on the list will not be considered for scheduling. The channel estimates for each selected terminal are retrieved in step 1044. The hypotheses of this hypothesis formed by Nτ via the selected terminal will be evaluated, and the corresponding vector hypid of the post-processing SNR for each sub-prediction is derived at step 1046. The best sub-hypothesis is selected and the data transmission rate corresponding to the best hypothesis S N R will be determined in step 1048. Again, the schedule and data transfer rate can be reported to the terminal in the hypothesis. Metrics and Systems for Terminals in the List -130 - 1269549 Code Transfer - Continued (126) The metric is then updated in step 1050. In a particular embodiment, the 'best sub-hypothesis may correspond to the normalized priority order that is closest to the terminal after its metric is updated.

The uplink schedule of the system is further detailed in U.S. Patent Application Serial No. 09/859,346, the entire disclosure of which is incorporated herein by reference. Methods and Apparatus for Resources, which are assigned to the assignee of the present invention and incorporated herein by reference.

For a scheduling scenario in which the terminal is selected according to the priority of the terminal and scheduled for transmission (downlink or uplink), a bad terminal cluster may occasionally occur. A "bad" terminal unit means that it causes a strong linear dependence in the assumed channel response matrix, resulting in a lower overall throughput for each terminal in the group. When it happens, the terminal can be substantially unchanged in several frames. In this mode, the scheduler may not be able to get rid of this particular terminal until the change in the priority of the terminal is sufficient to change the eligibility in that group. To avoid the above-mentioned "clustering" effect, the scheduler can be designed to confirm this condition before the designated terminal is available to the available transmission channel, and/or to detect its condition in the event of its occurrence. Many different techniques can be used to determine the degree to which the matrix iL is linearly dependent. These techniques include solving the eigenvalues of ii, and solving the SNR of the post-processing signal, using continuous cancellation receiver processing techniques or linear space equalization techniques and others. Detecting this cluster condition is usually easy to implement. In the event that the cluster condition is detected, the scheduler can rearrange the order of the terminals (e.g., in a random manner) in an attempt to reduce the linear dependence within the matrix. One -131 - (127) 1269549

The Walker case was also designed to force the scheduler to select the end unit that caused the "good" hypothetical matrix EL (i.e., with minimal linear dependence). VI. Performance

吏: The techniques described herein provide improved system performance (e.g., higher pass). Simulations have been implemented to quantify improvements in system throughput that may be due to some of these technologies. In the simulation, the wanted-reaction matrix of the array of coupled transmit antennas and the receive antenna is assumed to consist of an equal-variation, zero-mean Gaussian random variable (ie, "independent composite Gaussian hypothesis"). The average flux used to randomly select NT (1 x Nr) channels is evaluated. Please note that the flux is 5〇% of the channel i, as determined by the theoretical capacity limit of shann〇n.

Figure 11A shows the average downlink throughput for a terminal of a mim〇 system with 4 transmit antennas (ie Ντ = 4) and 4 receive antennas (ie heart = 4) for single-user mode and multiple uses. MIM〇 mode (ie N-SIMO mode). The analog flux associated with each mode of operation has a function to average the post-process SNR. The average throughput of the single user MIM mode is not shown in Figure 1110, and the average throughput of the multiple user mode is shown in Figure 1112. As shown in Figure 1 Α, the analog flux associated with the multi-user ΜΙΜΟ mode specified using the maximum-maximum standard antenna shows better performance than the single-user ΜΙΜΟ mode. In single user mode, the terminal advantageously benefits from the use of continuous cancellation receiver processing to achieve higher post processing SNR. In the multi-user mode, the scheduling scheme can use multiple user-selective diversity to achieve improved performance (ie, higher throughput), even if each terminal uses linear space (eg, MMSE) processing techniques -132 - 1269549

(128) surgery. In fact, the average downlink throughput caused by multiple user diversity in the multiple user mode exceeds the division by one transmission interval, four equal partial sub-slots, and each terminal is assigned to its own sub-slot. The balance achieved. The scheduling scheme used in simulating the single-user and multi-user authentication modes is not designed to provide appropriate fairness, and some terminals will observe higher average throughput than other terminals. When a fairness criterion is adopted, the flux difference for the second mode of operation may be reduced. However, the ability to reconcile both single-user and multi-user modes provides additional flexibility in wireless data services. Figure 1 1 shows the average uplink pass flux associated with 4 receive antennas (ie N r = 4) with various numbers of single antenna terminals (ie NT = 1, 2 and 4) with respect to an independent composite Gaussian assumption - Limit the environment (ie the interference power system is much larger than the thermal noise power). 4 The transmission antenna (ie Ντ = 4) is much larger than the 1 transmission antenna (ie Ντ = ,, the gain increases with SNR. When the SNR is extremely high, the capacity of Ντ == 4 is nearly four times Ντ = 1. In the case of very low SNR, the gain between these two cases is reduced and negligible. In low or no interference environments (eg thermal noise - limiting), the flux of τ = 4 cases is even It is more than shown in Fig. 11A. In the thermal noise/restricted environment, the interference power system is lower (for example, close to zero) and the SNR is 6 decibels, which is substantially larger than the case of Ντ = 4 in Fig. 1 . In the example, when a single terminal receives at SNR of 10 dB, the average throughput achieved by the terminal is 2,58 bps/Hz. When the 4 terminals are allowed to transmit simultaneously, the overall flux is equal to Ντ = 4 SNR = 1〇 decibel in the curve + 10.1og10(4)=

• 133 - 1269549 fiber (129) 16 knives. Therefore, in the thermal noise/restricted environment, the overall throughput of the 4-terminal is transmitted at 8,68 bps/Hz or nearly 3.4 times that of a single terminal.

In an interference-limiting system (e.g., a cellular network), the throughput provided by each cell with multiple S I Μ 0 transmissions in conjunction with the continuous cancellation receiver processing at the base station is selected for the terminal set point. a function. For example, at an SNR of 10 dB, when four 1x4 SIMO terminals allow simultaneous transmission, the capacity will be more than twice. At a 20 dB SNR, the capacity is increased by a factor of 26 compared to a single 1x4 terminal. However, a higher operating set point usually means a reuse factor for a larger frequency. That is, fragments of cells that use the same frequency channel at the same time may need to be reduced to achieve the demand S N R corresponding to a higher operational set point, which may then reduce the overall spectral efficiency (measured in bp s/Hz/cell). In maximizing the network capacity of this solution, there will therefore be a measure of the choice between the particular operational set point and the demand frequency reuse factor.

Figure 1 1 c shows the cell flux for a simulated network of one cell, with other Ντ = 1, 2 and 4 simultaneous terminals. Each cell receives the antenna using Nr = 4. All terminal systems are power controlled to achieve a given set point. The test shows that there is a range of SNR setpoints, where the cell flux for the Ντ = 4 terminal is twice as large as that allowed for only one single terminal. The components of the transmitter and receiver unit can be implemented with one or more digital signal processors (DSPs), dedicated integrated circuits (ASICs), processors. Microprocessor, controller, microcontroller, field programmable gate array (FPGA), programmable logic device 'other electronic units, or any combination thereof. Some of the functions and processes described herein can also be implemented using software implemented on a processor -134-1269549 (130).

Certain features of the invention may also be practiced using a combination of software and hardware. For example, spatial processing, spatial time processing, continuous cancellation receiver processing, full-CSI processing, CSI derivation (eg, channel SNR), scheduling, etc. may be based on the processor (controller 230 and/or in FIGS. 2A and 2B) 270) The code is executed and implemented. The software code can be stored in a memory unit (e.g., memory 2 3 2 or 2 7 2 ) and implemented by the processor (e.g., power control processor 230 or 270). The memory unit can be implemented within the processor or external to the processor, in which case the memory unit can be communicatively coupled to the processor via various means known in the art. The headings included here for reference are used to assist in locating certain segments. These headings are not intended to limit the scope of the concepts described herein, and these concepts can be applied to other paragraphs throughout the document.

The previous description of the specific embodiments of the present invention is provided to enable any person skilled in the art to make or use the invention. The various modifications of the specific embodiments are apparent to those skilled in the art, and the general principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not limited to the specific embodiments shown herein, but rather the broadest exemplary representation of the present invention and the novel features. The communication system 102a coverage area 102b coverage area - 135 - 1269549 (131) 102c Coverage area 102d Coverage area 102e Coverage area 102f Coverage area l〇2g Coverage area 104a Base station 104b Base station 106a Base station 106 Terminal 106b Terminal 106c Terminal 106d Terminal 106e Terminal 106f Terminal 106g Terminal 132 Memory 208 Data Source 210 Data Processor 210d Data Processor 2 lOx Data Processor 220 ΤΧ ΜΙΜΟ Processor 220a ΤΧ ΜΙΜΟ Processor 220d ΤΧ ΜΙΜΟ Processor 222 Modulators

-136 - 1269549 (132) 224 Antenna 230 Controller 234 Scheduler 252 Antenna 254 Demodulator 260 RX ΜΙΜΟ Data Processor 260e RX ΜΙΜΟ Data Processor 270 Controller 1 272 Memory 278 Source 280 ΤΧ Processor 282 ΤΧ Data processor 300 transmitter 300a transmitter 300b transmitter 300c transmitter 300d transmitter 300e transmitter 312 encoder 314 interleaver 316 mapping unit 318 weighting element 322 channel ΜΙΜΟ processor 324 demultiplexer

-137- 1269549 绔说说_Continued (133) 326 Combiner 330 Data Processor 340 Inverse Fast Fourier Transformer 342 Cyclic Header Generator 344 Upconverter 410 Space/Space Time Processor 410c Space/Space Time Processing 410e spatial/spatial time processor 412 filter 414 multiplier 416 combiner 420 matrix processor 422 multiplier 424 multiplier 428 adaptive processor 432 forward receive processor 434 adder 436 combiner 438 channel Data processor 440 feedback processor 448 CSI processor 450 receive processing stage 450 first stage 450η final stage -138-1269549

(134) 460 Channel Data Processing 460a Channel Data Processing 470 Interference Canceller 470a Interference Canceller 472 Channel Simulator 480 RX Processor 484 Demodulation Element 486 Deinterleaver 488 Decoder 730 Library

-139

Claims (1)

Or a plurality of selected; and modulation 2. 1269549 pick, patent scope K - a method for transmitting data in a multi-directional proximity multi-output system, comprising: selecting one or more for data transmission Terminals; receiving channel state information (CSI) indications of the selected one or more selected terminals; processing the ~ greedy according to the receiving CSI to provide a plurality of modulated signals via a plurality of transmitting antennas One or more selected terminals. For example, the method of claim 1 wherein the system transmits data via a plurality of modes of operation. 3. The method of claim 2, wherein the duplex comprises a single user mode, characterized in that the plurality of transmit antennas are used for data transfer to a single terminal having a plurality of bits. 4. The method of claim 3, wherein the transmitting in the ΜΙΜΟ mode data to the single terminal comprises a plurality of data streams transmitted on the modulated signal. 5. The method of claim 2, wherein the duplex comprises a multi-user mode, characterized in that a plurality of transmit antennas are used for data transfer to a plurality of terminals having a common complex. 6. The method of claim 5, wherein the signal of the channel terminal of a modulation (ΜΙΜΟ) communication condition is transmitted to a single user that can be configured to use the plurality of receiving antennas in a plurality of operating modes The plurality of operational modes use the plurality of receiving day signal systems design 1269549 Each of the plurality of terminals of the multiple user mode is used. 7. The method of claim 2, wherein the plurality of modes of operation comprise a hybrid mode characterized by using the plurality of transmit antennas for data transfer to a combination of a SIMO and a terminal, wherein A modulation signal system is designed for each S I Μ terminal, and the multi-modulation signal system is designed for use in each terminal. 8. The method of claim 2, wherein the plurality of modes of operation comprise a diversity mode, characterized by using the plurality of transmit antennas for reliably transmitting a single data stream to a plurality of receive antennas A single terminal. 9. The method of claim 2, wherein the plurality of modes of operation comprise a pass diversity mode, characterized in that the plurality of transmit antennas are used for data transmission to a single terminal having a single receive antenna machine. 10. The method of claim 1, wherein the terminal for data transmission is selected based on an estimated interference-to-interference ratio (SNR) achieved by the plurality of transmitting antennas. 11. The method of claim 1, wherein the SNR is derived at the terminal based on a preamble included in the plurality of modulated signals. 12. The method of claim i, wherein the terminal for data transmission is selected based on a radio frequency characteristic of a channel formed by a plurality of transmit antennas and a plurality of receive antennas of the terminal. 1269549 _ 申利利转宁 Continued page 13. The method of claim 12, wherein the radio frequency characteristic is derived at the terminals based on the preamble signals contained in the plurality of modulated signals . 14. The method of claim 1, further comprising: assigning a plurality of transmit antennas to the one or more selected terminals in accordance with the receive Cs. 15. The method of claim 1, further comprising: assigning each selected terminal to the one or more transmit antennas. 16. The method of claim 1, wherein the terminal for data transmission is selected based on one or more metrics. 17. The method of claim 16, wherein one of the one or more metrics is indicative of a throughput achievable by the selected terminal. 18. The method of claim 16, wherein one of the one or more metrics is a function based on the SNR achieved by the selected terminal. 19. The method of claim 1, wherein the terminal for data transmission is selected according to the order of priority of the terminal. 20. The method of claim 19, wherein the priority of a particular terminal is determined by an average throughput of the terminal. 21. The method of claim 1, wherein the processing comprises: encoding and modulating data of the one or more selected terminals in accordance with the receiving C S I . 22. The method of claim 10, further comprising: at the terminal, based on the S NR estimated by one of the modulated signals, compiling 1269549 _ Ipl patent traveler continuation page and modulating each modulated signal data of. 23. The method of claim 12, further comprising: pre-adjusting the modulation symbol based on a matrix of eigenvectors comprised of the radio frequency features of the one or more selected terminals. 24. The method of claim 1, wherein the processing comprises: adjusting a data transmission rate of the one or more selected terminals in accordance with the receiving C S I . 25. The method of claim 1, further comprising: receiving feedback from the one or more selected terminals; and adjusting at least one characteristic of the modulated signals based on the received feedback. 26. The method of claim 25, wherein the transmit power of the modulated signal is adjusted in accordance with the receive feedback. 27. The method of claim 25, wherein the data transmission rate of the modulated signals is adjusted in accordance with the received feedback. 28. The method of claim 25, wherein the data encoding and modulation of the modulated signals are adjusted in accordance with the received feedback. 29. The method of claim 1, wherein the plurality of modulated signals are transmitted at a power level determined in part by one of a maximum allowable power level or a plurality of power backoff factors. 30. The method of claim 29, wherein the one or more power concession factors are selected to reduce interference with Zheng near cells. 31. The method of claim 29, wherein the one or more power backoff factors are selected based on system load. 1269549 32. The method of claim 29, wherein the one or more power backoff factors are selected based on performance achievable by the terminal in the system. 33. The method of claim 1 wherein the C S 1 includes an estimated interference to interference ratio (SNR) for a plurality of transmission channels for data transmission. 34. The method of claim 1, wherein the C S I includes an indication of a data transmission rate supported by a plurality of transmission channels for data transmission. 35. The method of claim 3, wherein the SNR is derived at the terminal machine in accordance with spatial processing. 36. The method of claim 35, wherein the spatial processing at a terminal comprises a channel correlation matrix inversion (CCMI) technique or a minimum mean square error (MMSE) technique. 37. The method of claim 3, wherein the SNR is derived at the terminal according to spatial time processing. 38. The method of claim 3, wherein the spatial time processing comprises an MMSE linear equalizer (MMSE-LE) technique or a decision feedback equalizer (DFE) technique. 39. The method of claim 3, wherein the SNR is derived at the terminal based on continuous cancellation receiver processing. 4. The method of claim 1, wherein the system implements orthogonal frequency division multiplexing (OFDM). 41. The method of claim 1, wherein the system implements a code division 1269549 Proximity (CDMA). 42. A method for transmitting data on a downlink in a multi-directional proximity multiple input multiple output (mim〇) communication system, comprising: receiving a plurality of transmit antennas at a plurality of terminals Estimating a noise to interference ratio (SNR); selecting one or more terminals for data transmission based on the estimated SNR; processing data of the one or more selected terminals based on the estimated SNR to provide a plurality of tones Transmitting the plurality of modulated signals to the one or more selected terminals via a plurality of transmit antennas, and wherein the system is configurable to transmit data via a plurality of modes of operation, the mode comprising Single user mode, one multi-user mode and one hybrid mode. 43. A method for transmitting data in a multi-directional proximity multiple input multiple output (μιμ〇) communication system, comprising: channel state information (csI) for receiving channel conditions of a plurality of terminals; Transmitting one or more terminals of the uplink data transmission to the one or more selected terminals of the one or more selected terminals; receiving the one or more selected terminals from the plurality of receiving antennas And a plurality of modulated signals; and processing the plurality of received signals to recover the data transmitted by the terminal device of the one or more selected terminals. 44. The method of claim 4, wherein the terminal for data transmission is selected based on the estimated interference-to-interference ratio (SNR) of the plurality of available transmission channels. 45. The method of claim 4, wherein the terminal for data transmission is selected based on a radio frequency characteristic of a channel formed by a plurality of transmitting antennas and a plurality of receiving antennas of the terminals. 46. The method of claim 4, wherein the terminal for data transmission is selected in part based on one of the maximum allowable power levels or a plurality of power backoff factors. 47. The method of claim 44, wherein the SNR is derived from spatial processing. 48. The method of claim 44, wherein the SNR is derived from spatial time processing. 49. The method of claim 44, wherein the SNR is derived from continuous cancellation receiver processing. 50. A base station in a multi-directional proximity multiple input multiple output (ΜΙΜΟ) communication system, comprising: a scheduler operative to select one or more terminals for data transmission; a controller, Operating in a channel status information (CSI) indication for receiving channel conditions of the one or more selected terminals, and providing one or more control items in accordance with the receiving CSI; a data processor operating in accordance with the One or more controls 1269549 _ _ _ patent continuation page items to process the data of the one or more selected terminals to provide a plurality of modulated symbol streams; a modulator that operates to generate a plurality of And a plurality of transmit antennas configured to transmit the modulated signals to the one or more selected terminals. 51. A base station in a multi-directional proximity multiple input multiple output (ΜΙΜΟ) communication system, comprising: a component for selecting one or more terminals for data transmission; a component for receiving channel status information (CSI) a channel status information (CSI) indication for receiving channel conditions of the one or more selected terminals, and providing one or more control items according to the receiving CSI; means for processing data for use in accordance with the one or And a plurality of control items to process data of the one or more selected terminals to provide a plurality of modulated symbol streams; generating a plurality of components of the plurality of modulated symbol streams; and transmitting the same Signaling to components of the one or more selected terminals. 52. A terminal in a multi-directional proximity multiple input multiple output (MIMO) communication system, comprising: at least one front end processor operative to receive and process at least one received signal to provide a received modulated symbol ; an RX ΜΙΜΟ / data processor, which operates to receive 1269549 Processor processing techniques for receiving and processing the received modulation symbols to provide estimated modulation symbols in the transmitted signals, wherein the RX/data processor is further operative to provide channels for the plurality of transmitted signals Conditional Channel Status Information (CSI) indication; and a data processor configured to receive and process the C SI for transmission from the terminal. 53. A terminal in a multi-directional proximity multiple input multiple output (ΜΙΜΟ) communication system, comprising: means for processing at least one received signal to provide received modulated symbols; processing the received A component of a modulation symbol that is based on a receiver processing technique to provide an estimated modulation symbol in the transmitted signals; a component that derives channel state information (CSI) indications of the channel conditions of the plurality of transmitted signals; And processing the components of the CSI for transmission from the terminal.